Adverse Childhood Experiences and Intimate Partner Violence Among Recent Adults: Mental Health and Race

Abstract

Previous research has linked childhood exposure to family dysfunction and violence with long-term health outcomes, including mental health and later exposure to violence. However, fewer studies have modeled depressive symptoms (stemming from childhood adversity) as a key linking variable with later intimate partner violence (IPV) – particularly among diverse youth. The present study investigated (a) the direct effect of adverse childhood events (ACEs) on intimate partner violence, (b) the indirect effect of ACEs on intimate partner violence through depressive symptoms, and (c) the moderation of these associations by race. Method: Data were drawn from 702 participants (80% female) 18 years of age from panel data collected in 2019 from the 18 & Life Project. Participants self-reported adverse childhood experiences, intimate partner violence, and depressive symptoms using multiple items for each measure. Additional demographic variables assessed included gender, race, and sexual orientation. Structural equation modeling using Mplus Version 8 software was used to test hypothesized associations. Results: While results indicate a direct association between ACEs and IPV. Evidence for an indirect link through depressive symptoms was only found for White youth. The findings of this study provide evidence for the persistent long-term influence of ACEs on mental health and relationship experiences in later life. Conclusions: Our results suggest a need for early family and community-level interventions to protect children from exposure to early adverse experiences in order to protect their mental health and ensure successful relationship experiences.

Introduction

Intimate partner violence (IPV) is a highly prevalent problem in the United States, affecting millions of individuals (CDC, 2022a, b). Nearly 25% of women and 10% of men in the United States have experienced IPV (sexual, physical, or stalking), while over 80 million have experienced psychological aggression by an intimate partner. This is a global problem, as approximately 30% of all women have been subjected to intimate partner violence (WHO, 2021), approximately 10% of this IPV takes place in adolescent romantic relationships. The prevalence rates of intimate partner violence take on greater importance in consideration of the fact that nearly 10% of homicides occur between intimate partners (FBI, 2015). Key adversities experienced during childhood and adolescence further compound the risk for IPV throughout adulthood (Halpern et al., 2009). The present study focuses on the specific and poignant age of 18 as a window of time between childhood and adulthood when early or recently experienced adversities may already be increasing the likelihood of relationship violence potentially setting youth on a trajectory toward future or chronic IPV experiences (Shen & Kusunoki, 2019; Smith et al., 2003; Widom, 1999). Specifically, the present study examines the association between ACEs and IPV, via depressive symptomology, among 18-year-old men and women. Prior research on these associations have not included a representative cohort of young women, who are at unique risk for both ACEs and IPV, especially when considering racial/ethnic influences (Vezena & Hebert, 2007).

Adverse Childhood Experiences (ACEs) and IPV

Adverse childhood experiences (ACEs) include a cluster of experiences children may encounter in their home that range from difficult to traumatic (Finkelhor, 2020) and influence an array of developmental and health outcomes over time (CDC, 2022a, b; Monnat & Chandler, 2015). Research has shown that ACEs increase IPV among young adults, as well as lifetime aggressive behavior (Forster et al., 2022; King, 2021). This influence on IPV may largely be a result of early adversity on youths’ role modeling, stemming from multiple forms of household dysfunction captured by ACE measures (e.g., parent mental illness, substance abuse). This is consistent with the developmental–interactional perspective suggesting the experience of parental aggression and dysfunction may transmit to offspring’s intimate relationships (Capaldi & Gorman-Smith, 2003; Conger et al., 2013; Cui et al., 2013). Maladaptive stress-response behaviors stemming from early adversity, abuse, or neglect (especially behavioral aggression and poor mental health) have complex links with intimate partner violence increasing risks of both victimization and perpetration (Mair et al., 2012).

ACEs Increase Depressive Symptoms

Early childhood experiences have been shown to be indirectly associated with IPV through poorer mental health among youth` (Terrell et al., 2022). Previous studies have shown that ACEs increase depressive symptoms in youths (Cheong et al., 2017; Mair et al., 2012). Youths whose parents have experienced mental health problems and youths who have experienced sexual abuse are at the greatest risk of developing depressive symptoms later in life (Giano et al., 2021). Several aspects of ACEs—such as parental separation/divorce, household dysfunction, and exposure to violence have been shown to be stressful for youths and associated with elevated depressive symptoms (Elmore & Crouch, 2020). Research has shown a graded dose–response relationship between ACE exposure and psychological and behavioral problems (Merrick et al., 2017). Exposure to four or more measured adversities increase the likelihood to experience depression and anxiety disorders in adulthood, and severe ACE exposure has also been linked to inflammation (CRP-C Reactive Protein) and neuronal health (BDNF-Brain Derived Neurotrophic Factor) in adulthood (Hughes et al., 2017; Watt et al., 2020).

Depressive Symptoms Increase IPV

A significant association exists between poor mental health and IPV (Black, 2011; MMWR, 2014). Greater depressive symptoms have been associated with an increase in the perpetration of physical and psychological IPV among both men and women (Barros-Gomes et al., 2016). Furthermore, the influence of depressive symptoms on physical IPV was found to occur indirectly through psychological IPV, with a dyadic indirect effect occurring such that female depressive symptoms influenced male psychological IPV and in turn physical IPV (and vice versa; Barros-Gomes et al., 2016).

IPV Disrupts Life

The impact of IPV has health, education, and economic consequences (Gibbs et al., 2018; Hess & Del Rosario, 2018). Health-related quality of life is significantly lower among female IPV victims and male and female both perpetrators and victims (Costa et al., 2014). IPV has been shown to detrimentally contribute to ineffective parenting among mothers (Pels et al., 2015), which aligns with research on the intergenerational transmission of childhood exposure to adversity and violence.

Moderation by Race

Consistent with the minority stress perspective, stress responses are dependent upon culture and social context (e.g., racism) that are largely determined by race/ethnicity and minority status (Dentato, 2012; Meyer, 1995; Williams et al., 2003). Minority stress gives rise to conflict with the social environmental experience due to values between minority and dominant individuals. Youths who experience such conflictual social contexts may be less vulnerable to adverse family experiences (i.e., ACEs) because they may develop hardiness toward stressors through extra-familial experiences (steeling effect) (Rutter, 2012; Wickrama et al., 2020). Thus, the associations between ACEs, mental health, and IPV may be contextualized or moderated by race/ethnicity, resulting in differences between these associations (Alvarez et al., 2019).

The Present Study

The present study examined the direct effect of ACEs on intimate partner violence among contemporary and representative sample of 18-year-old women in the United States (n = 702). Many prior studies have utilized secondary or post-secondary school samples; however, early life adversities, poor mental health, and exposure to violence all increase the likelihood that women with higher ACEs and experiences of IPV will not be found in these settings (Vezena & Hebert, 2007). This study also tested the mediating role of depressive symptoms in the association between ACEs and intimate partner violence among young women beginning the transition to adulthood. Furthermore, this research examined differences in the associations and prevalence of ACE and IPV by race (Pro et al., 2020).

The Hypothesized Model

ACEs were hypothesized to increase intimate partner violence (Path A, Fig. 1). This direct association was further hypothesized to be mediated by depressive symptoms indirectly linking ACEs and IPV. Consistent with the stress process perspective (Pearlin et al., 1981), greater ACEs (which are stressful for youths) were hypothesized to lead to increased depressive symptoms in youths (Path B, Fig. 1). Elevated depressive symptoms were hypothesized to increase experiences of intimate partner violence in young adulthood (Path C, Fig. 1; Woodin et al., 2013). These direct and mediating effects were hypothesized to be moderated by race. The predicted associations were hypothesized to occur while controlling for gender and sexuality, as childhood adversity and stress experiences may be influenced by these sociodemographic factors (Schnarrs et al., 2019).

Fig. 1.

The Theoretical Model:

Method

Participants and Procedure

Following university IRB approval, a representative sample of US 18-year-olds (n = 702) was obtained from the Qualtrics company. In collaboration with research marketing firms, Qualtrics recruits participants nationally from panel sources, based on the inclusion criteria of the study. Participants self-reported their ACEs, social media use, cyberbullying, IPV, mental health, and other demographic variables. Qualtrics panel data is designed to mirror the census representation selected by researchers and has been shown to be as credible and provide similar data quality as traditionally collected samples (Walter et al., 2019). Qualtrics panel data is the most demographically representative and has the most accurate psychometric properties of all the online panel data collection systems (Boas et al., 2018). A total of 20% of the sample consisted of males while 80% of the sample consisted of females. 12% of the sample is Black, 69% White, 8% Hispanic, 2% Native American, 5% Asian, and 3% reported identifying with more than one racial group.

Measures

Adverse Childhood Experiences (ACEs)

Participants responded to 10 items corresponding to the ACEs measure as described by the Centers for Disease Control and Prevention (Schulman & Maul, 2019; US Department of Health and Human Services). The ACEs measure was developed and used by CDC for children and youth in the United States and has shown to be valid (predictive, content, convergent, divergent, and discriminant), both in the US and globally (Felitti et al., 1998; Anda et al., 2006; Chapman et al., 2004; Hills et al., 2000, 2001; Dietz et al., 1999; Lacey & Minnis, 2020). Items asked included whether the adolescent experienced the following while growing up: sexual abuse, physical abuse, verbal abuse, neglect in love and care, neglect in providing material needs, divorced/separated parents, mother abused, alcohol or drug use in the household, mental illness or suicide in the household, and household members experience with imprisonment. Responses options were (0 = no; 1 = yes). Cronbach’s alpha for the measure was 0.81.

Intimate Partner Violence (IPV)

is a summed composite of 3 items. The questions asked how often each participant experienced romantic partner relationship abuse (physical abuse, emotional abuse, sexual abuse) in the last year. These items were measured on a scale from 0 to 6 with 0 = this has never happened or not in the past year, 1 = once in the past year, 2 = twice in the past year, 3 = 3–5 times in the past year, 4 = 6–10 times in the past year, 5 = 11–20 times in the past year, 6 = more than 20 times in the past year. The three items are used as indicators of a latent IPV construct. (Cronbach’s α = 0.78). Nationally representative studies using similar IPV scales on adolescents have shown good validity of the scale (Rothman et al., 2022). An IPV measure with similar items derived from the PROMIS measure has been psychometrically validated (PROMIS, 2023).

Depressive Symptoms

consisted of 6 items from the Center for Epidemiological Studies Depression scale (CESD) (Radloff, 1977) as available from the PROMIS scale included in the questionnaire. The CES-D is a depressive symptoms scale widely used across the United States and globally for its validity and adaptability in diverse contexts (Fu, 2020). The present measure (a short version of CES-D) has been used in large representative national studies such as Add Health showing good validity (Shaffer, 2014). These items assessed whether in the past week the participant felt sad, thought their life had been a failure, felt depressed, was unhappy, felt hopeless about the future, and felt fearful, on a scale from 0 (rarely or none of the time) to 3 (all of the time) (Cronbach’s α = 0.91).

Demographic Variables

Gender (1 = Male; 2 = Female; 3 = Other). Race was measured using self-reports from participants. A set of two dichotomous variables (coded 0 and 1) was used to contrast race category Black against the reference group, White. Sexual orientation (0 = not heterosexual; 1 = heterosexual,) was self-reported by participants.

Statistical analyses

The analyses were conducted using Mplus Version 8 software (Muthen & Muthen, 2017). Preliminary analysis consisting of frequencies, descriptive statistics, and variable diagnostics was conducted. Structural Equation Modeling was used to test the hypothesized associations. ACE, Depressive symptoms and IPV were defined as latent variables (factors) using observed measures as reflective indicators. Factor loadings reflect the validities of constituent indicators in relation to the latent variable. The hypothesized associations between latent variables were estimated. Statistical significance of each of estimated regression coefficients was tested using z values (coefficient/standard error) and tested (Bollen, 1989).

To assess the overall model-fit, we used the comparative fit index (CFI > 0.95) and root mean square error of approximation (RMSEA < 0.08), and the Chi-Square/df ratio to evaluate model fit (Hu & Bentler, 1999). Bootstrap procedures were used test the indirect effects with 2.5% above and below the estimate to create the 95% Confidence Interval cut points.

Analyses included 569 participants as only those that reported a race of White or Black were included in the analyses. Other race/ethnicities (Native American, Asian American, Hispanic American) were excluded due to small sample sizes. Full Information Maximum Likelihood (FIML) procedure, which estimates model parameters and standard errors from all available data (Enders & Bandalos, 2001) is used in the analysis.

Goodness-fit indices (Chi-Square/df, RMSEA, and CFI) are presented in Figs. 2, 3 and 4, as well as in the text. The mediation model in Fig. 2 indicates Chi-Square/df: 3.54, RMSEA 0.06, and CFI 0.90. The model fit values in Fig. 3 indicates Chi-Square/df: 1.68, RMSEA 0.08, and CFI 0.78. The mediation model in Fig. 4 indicates Chi-Square/df: 2.85, RMSEA: 0.06, CFI: 0.90. These values indicate that the data is a good fit with the model; most or all of the model fit values in each figure are in the acceptable range, RMSEA between 0.05 and 0.08, CFI 0.90 and greater, chi-square/df ratio less than 5 (Hu & Bentler, 1999; Kim et al., 2016; West et al., 2012). Thus, the interpretation of the coefficients in the models is valid.

Fig. 2.

Mediation Model, Full Sample

Fig. 3.

Mediation Model for Black Adolescents

Fig. 4.

Mediation Model for White Adolescents

Results

Table 1 shows bivariate correlations among study variables as well as mean values and standard deviations.

Table 1.

Bivariate correlations among study variables

*p < .05, **p < .01

Testing the Hypothesized Model

The results are presented in Fig. 2. All the factor loadings of indicators of ACE, and IPV are significant (p < 0.05). Figure 2 also shows the significant direct and indirect effects of ACEs on intimate partner violence (IPV) for all adolescents in general. The direct effect of ACEs on IPV suggests that early childhood adversity has a long-term influence on relationship experiences (β = 0.15, p < 0.05). This may partly reflect intergenerational transmission of relationship behavior consistent with the developmental-interactional perspective. The significant indirect mediational effect (indirect effect β = 0.05 p < 0.05); from ACEs to IPV through depressive symptoms (ACEs to depression β = 0.48, P < 0.001 and depression to IPV β = 0.11, p < 0.05) suggests a psychological processes responsible for this association. Furthermore, both the direct and the indirect effects are simultaneously significant. The significant effect of depressive symptoms on IPV confirms the previous findings of vulnerability to IPV among those with depressive symptoms and greater likelihood of victimization.

Testing Moderating Effect of Race: Modeling for Black Adolescents

We tested race moderation of the above observed association by estimating the model for Black sub-sample and comparing results with those of total sample (Fig. 2) and White sample (Fig. 3). Figure 3 shows the results of the study model for Black adolescents. All the factor loadings of indicators of ACE, and IPV are significant (p < 0.05). There are significant differences between the total sample of adolescents and Black adolescents in the paths. The direct effect from ACEs to IPV is not significant (β = 0.00, p > 0.05). Moreover, the mediation path is not significant (indirect effect β = -0.10 p > 0.05). The influence from ACEs to depressive symptoms is significant (β = 0.63, p < 0.001), while the influence of depressive symptoms on IPV is not significant (β = -0.27, p > 0.05). The stark difference in the significant paths from the general population of adolescents may be explained by the minority stress perspective. Black youths who are exposed to adverse social experiences in the larger society (e.g., discrimination) may not be susceptible to adverse family experiences (high hardiness) and do not show behavioral consequences (e.g., IPV) consistent with developmental-interactional perspective. In addition, although Black youths are psychologically vulnerable to ACEs showing elevated depressive symptoms, the association between depressive symptoms and IPV is not significant. This may be due to less variation in elevated depressive symptoms in Black youths.

Testing Moderating Effect of Race: Modeling for White Adolescents

Figure 4 depicts the study model results for White adolescents. All the factor loadings of indicators of ACE, and IPV are significant (p < 0.05). In contrast to the Black adolescent model results, this shows the significant mediation effect from ACEs to IPV through depressive symptoms (indirect effect β = 0.08 p < 0.01); from ACEs to depression (β = 0.47, p < 0.001) and from depression to IPV (β = 0.17, p < 0.01). These results support the psychological processes stemming from childhood adverse experiences leading to IPV. As in the results for the full sample of adolescents, for White adolescents the significant effect of depressive symptoms on IPV confirms the previous findings of vulnerability to IPV among those with depressive symptoms and greater likelihood of perpetration. However, the influence of ACEs on IPV did not reach significance for White youths (β = 0.15, p > 0.05). This may be due to the reduced sample size. Table 2 shows the prevalence of ACEs, IPV, and depressive symptoms among all adolescents. ACEs, consisting of the 10 items belonging to the subcategories of abuse, neglect, dysfunctional household, and divorce or separated parents showed varying levels of prevalence rates. White adolescents had a higher prevalence of depressive symptoms with 16.6% than Black adolescents. Male adolescents had higher levels of Physical IPV with 11.06% than female adolescents. Female adolescents had higher levels of ACEs Abuse with 18.3% than male adolescents. Heterosexual adolescents had a higher level of ACEs Neglect with 16.4% than others. Non-heterosexual adolescents had higher levels of ACEs Abuse with 23.6%. than others.

Table 2.

Percentage with high prevalence (≥ mean + 1 SD). Calculated at sub-sample averages and standard deviations

Discussion

This study aimed to examine the influence of childhood experiences on relationship functioning and the role of mental health among youth just launching into adulthood. More specifically, this study tested the hypothesized mediation effect of depressive symptoms on the association between ACEs and intimate partner violence, building on previous work showing preliminary evidence for the hypothesized associations (Alvarez et al., 2019; Lee et al., 2020; Pro et al., 2020). The results provided evidence for these associations and extended beyond this aim to test these associations separately by racial group. The moderation tests for Black and White adolescents showed significant differences between Black and White adolescents in the associations of ACEs with depressive symptoms and depressive symptoms with IPV. The positive association between ACEs and IPV was significantly mediated by depressive symptoms in White adolescents but not in Black adolescents (steeling effect), suggesting a modifiable pathway.

The findings of this study support previous evidence for the persistent long-term influence of ACEs on mental health and relationship experiences in later life (Forster et al., 2022; Cheong et al., 2017). Findings also supported the developmental–interactional perspective, which includes the notion of intergenerational transmission of relationship behavior (Capaldi & Gorman-Smith, 2003; Conger et al., 2013; Cui et al., 2013). That is, early exposure to conflictual relationship dynamics among parents appears to be modeled by adolescents in their own relationship behaviors and experiences later in life. Moreover, the present findings support the notions of stress process theory (Pearlin et al., 1981). Accordingly, stressful early childhood experiences among adolescents have been shown to lead to elevated depressive symptoms through intermediate stress response mechanisms. For example, ACEs may directly generate depressive feelings such as a sense of hopelessness and helplessness in youths. The present study provided evidence for a significant link between greater depressive symptoms and increased exposure to intimate partner violence (Mair et al., 2012). This may be attributed to the fact that depressed individuals are typically more irritable and assertive, with decreased cognitive functioning, increasing their risk for violent behavior (Conger et al., 2010).

This study provided some support for the minority stress perspective. Accordingly, the results suggested that culture-specific and social context-dependent experiences might have led to the observed racial differences in the associations between ACEs, depressive symptoms, and intimate partner violence experiences (Dentato, 2012; Meyer, 1995; Williams et al., 2003). Minority youth, who often experience conflictual and discriminatory social interactions in the larger society, may develop hardiness toward adverse family experiences. Furthermore, minority adolescents may receive more social support from extended family members, who may be used as coping resources. Black adolescents experiencing minority stress may be less vulnerable to the effects of early childhood adversity, leading to differences between Black and White adolescents in the associations between ACEs, depressive symptoms, and IPV (Alvarez et al., 2019).

Alternative explanations may also exist warranting further study. For instance, it could be that other contextual factors (beyond ACEs and mental health) are more strongly linked to partner violence among black youth, such as poverty, which is disproportionately experienced by minority families (Gillum, 2019). In fact, the risk of IPV for black women has been found to be quite high and linked with a host of other contextual and interpersonal risks – and these current factors might play more influential roles in their interpersonal conflicts than childhood stressors (Gillum, 2021). Additionally, black youth (particularly women) may be less likely to report experiences of relationship violence, particularly if the perpetrator was a black male, in order to protect them from “a law enforcement system that has historically mistreated black men” (Gomez & Gobin, 2020). This may translate to overall lower reporting of IPV among black samples that masks true incidence rates (i.e., “culture of silence” regardless of partner’s race/ethnicity; Gomez & Gobin, 2020). So, there may be a paradox taking place where minority stress and discriminatory experiences both increase the risk of perpetrating violence and increase the likelihood that victims remain silent and underreport IPV (Lavner et al., 2018). However, this speculation requires further study.

Another notable limitation to this study is that the focus was intentionally on 18-year-old youth. The focus of IPV in this research was not limited to adolescence or adulthood – rather, the focus was on the narrow space in between to observe any immediate influences of ACEs on relationship violence as youth begin their transition to adulthood. Research has indicated that individuals can be trapped in a vicious loop involving IPV and mental health problems. That is, exposure to IPV appears to not only initiate mental health problems but also exacerbate existing mental health problems, thus making individuals more vulnerable to IPV (Mazza et al., 2021). Early life risks can exponentially compound the likelihood of these loops (Halpern et al., 2009). The bidirectional associations between IPV and mental health symptoms have been found longitudinally and across ethnic groups, establishing them as global health issues (Chatterji & Heise, 2021; Pakhomova et al., 2021). The present study tested the relationship between depressive symptoms and IPV but did not find significant bidirectional associations between these two constructs (at least not by the age of 18). Future research should further investigate potential bidirectional associations between depressive symptoms and IPV as these associations fluctuate during the transition to adulthood and, as stress responses to ACEs, can be related to violence victimization and perpetration (Mair et al., 2012).

Conclusion

Through the lens of stress process theory, implications of the present results can be extended to multiple levels of interventions: individual, family, and community social, health, and relationship intervention programs. Findings clearly indicate a need for early interventions to protect children from exposure to early adverse experiences (e.g., abuse, neglect, household dysfunction, and exposure to community stressors). By minimizing children’s exposure to ACEs or disrupting stress response behaviors and pathologies, early interventions can be protective for long-term mental health, minimize the cascading nature of risks (Wickrama et al., 2016), and increase the likelihood these youth will have successful relationship experiences in adulthood. While IPV was linked with ACEs in the current study, pre-existing risks are numerous (Halpern et al., 2009) and often dynamic (i.e., catalyzing cascading chains of risk that can take on a life of their own; Wickrama et al., 2016).

Individual interventions should not just focus on adolescents with depression symptomology, but specifically target depressed youth at the earliest launch into adulthood, who are more likely to be involved in relationship violence and at considerable risk for cyclical experiences of IPV across the life course (Smith et al., 2003).` Furthermore, community and school-level interventions can provide necessary social and mental health services and therapies for distressed adolescents. Recent research has emphasized the importance of support systems in identifying signs of and intervening in situations of abuse (Korkmaz et al., 2022) and, at the age of 18, many young women have heightened contact with these systems as they transition out of school and into other educational, workforce, or family situations. It is a sensitive window of opportunity for intervention that could potentially disrupt IPV loops throughout adulthood (Smith et al., 2003).

Declarations

Conflict of Interest Statement

On behalf of all authors, the corresponding author states that there is no conflict of interest.