Abstract
Black pregnant and postpartum individuals are at risk for intimate partner violence (IPV), and those with a history of childhood maltreatment and IPV are even more likely to be re-victimized during pregnancy. However, it is unknown if specific types of child maltreatment predict later IPV with and without a weapon better than others. The current study sought to (i) document the prevalence of childhood maltreatment and IPV and (ii) examine the relations among types of childhood maltreatment and later IPV with and without a weapon within a sample of Black individuals seeking prenatal care at a large public hospital in the southeastern United States. Participants (n = 186; mean age = 27.2 years, SD = 5.3) completed measures assessing childhood maltreatment and IPV with and without a weapon. Approximately 68.5% of participants (n = 124) endorsed experiencing childhood maltreatment, while 42.6% (n = 78) endorsed experiencing IPV. The bivariate relations among five childhood maltreatment types (i.e., sexual, physical, and emotional abuse, physical and emotional neglect) and IPV with and without a weapon were assessed. All childhood maltreatment subtype scores—except childhood physical neglect—were significantly higher among participants who reported a history of IPV with or without a weapon compared to participants who denied a history of IPV with or without a weapon. Logistic regression models revealed childhood sexual abuse emerged as the only significant predictor of experiencing IPV with a weapon (B = 0.10, p = .003) and IPV without a weapon (B = 0.11, p = .001). For every point increase in childhood sexual abuse subtype score, the odds of experiencing IPV with and without a weapon increased by 10% (OR = 1.10, 95%CI [1.04, 1.18]) and 12% (OR = 1.12, [1.05, 1.20]), respectively. Findings suggest that screening for childhood sexual abuse may provide a critical opportunity for maternity care providers to identify individuals at increased risk for IPV victimization with and without a weapon.
Introduction
Intimate partner violence (IPV), defined as physical or sexual violence, stalking, or psychological harm perpetrated by a current or former intimate partner, is recognized as a global health issue and social epidemic (Leemis et al., 2022; Niolon et al., 2017). IPV victimization is closely linked to a variety of adverse physical and mental health outcomes (Potter et al., 2020), including but not limited to acute injury, sexually transmitted diseases, chronic pain and inflammation, sleep disturbances, hypothalamic-pituitary-adrenal axis dysfunction, posttraumatic stress disorder (PTSD), depression, and anxiety (Dillon et al., 2013; Shen & Kusunoki, 2019).
Of concern, individuals who are socioeconomically disadvantaged, specifically those who belong to marginalized groups (e.g., racial/ethnic, gender, and sexual identity minorities) and/or have limited resources, experience IPV and its negative effects at disproportionate rates. Within the United States, BIPOC individuals (i.e., Black, Indigenous, and People of Color) are more likely to experience sexual, physical, and psychological IPV than White individuals (Black et al., 2011; Stockman et al., 2015). Further, compared to cisgender men, cisgender women and LGBTQIA+ individuals are more likely to experience injuries, fear, and posttraumatic stress symptoms as a consequence of IPV (Caldwell et al., 2012; Langenderfer-Magruder et al., 2014). Economic vulnerability is also related to increased risk for IPV (Breiding et al., 2017), at least in part because financial constraints can prevent those with fewer resources from leaving violent relationships (Bender et al., 2003; Matjasko et al., 2013). Given its myriad adverse effects, particularly among those who are socioeconomically vulnerable, researchers have sought to determine factors associated with IPV victimization within populations with intersecting marginalized identities.
Childhood Maltreatment as a Risk Factor for Later IPV Victimization
Childhood maltreatment, or exposure to abuse or neglect during childhood (i.e., physical, sexual, and emotional abuse or emotional and physical neglect before the age of 18), is an especially salient predictor of subsequent IPV victimization (Barrios et al. 2015; Chiang et al., 2018; Fereidooni et al., 2023; Krause-Utz et al., 2021). As such, extant literature has primarily focused on determining how different forms of childhood maltreatment predict later IPV victimization. Several studies have demonstrated that childhood sexual and physical abuse (CSA and CPA, respectively) are especially tied to later IPV victimization (Daigneault et al., 2009; Richards et al., 2017), while others suggest increased risk for later IPV victimization occurs across childhood maltreatment types (McMahon et al., 2015). More recently, studies have begun to examine how the type of childhood maltreatment differentially predicts later IPV victimization type (i.e., physical, sexual, or emotional IPV). For example, childhood maltreatment is associated with experiencing later physical and sexual IPV as well as threats of violence and stalking by intimate partners, even after controlling for sociodemographic variables (Papalia et al., 2021). However, to our knowledge, no studies have examined how different types of childhood maltreatment uniquely influence later IPV victimization with and without a weapon. The ability to predict IPV victimization with and without a weapon based on childhood maltreatment history would be instrumental in determining risk, as the perpetrator’s use of a weapon during IPV victimization tends to increase lethality and result in more deleterious psychological outcomes (J. C. Campbell et al., 2003; Sorenson, 2017).
Pregnant persons are particularly vulnerable to the poor outcomes associated with IPV. IPV victimization during pregnancy is related to poor maternal and fetal health outcomes, including the risk of spontaneous abortion and a lower infant birth rate (Alhusen et al., 2015; Bailey, 2010). While the literature is inconsistent about whether the overall prevalence of IPV changes during pregnancy—some individuals may experience a reduction in violence while others may experience more severe IPV (Agarwal et al., 2023; Bailey et al., 2010)— there is general consensus that those with experiences of IPV prior to pregnancy are at particular risk (Silva et al., 2011; Chisholm et al., 2017; Silva & Leite, 2020). Indeed, homicide is the leading cause of maternal mortality in the United States (J. Campbell et al., 2021; Cliffe et al., 2019; Deshpande et al., 2017; Wallace et al., 2021). Notably, most pregnancy-associated homicides involve a combination of intimate partner and gun violence (Lawn & Koenen, 2022), with two of the strongest risk factors being direct access to a gun and previous threats with a weapon (Spencer & Stith, 2020). Predicting risk for IPV with and without a weapon is especially critical among Black individuals seeking prenatal care from a public hospital setting, given their intersecting vulnerabilities and high rates of IPV, trauma-related concerns, and unmet treatment needs (Powers et al., 2020). A better understanding of risk among multiply marginalized care-seeking individuals would allow providers to approach routine care encounters as opportunities to enact preventive interventions designed to assess and mitigate IPV risk across the course of pregnancy and the postpartum period.
The Current Study
The current study sought to (a) document the prevalence of childhood maltreatment and IPV in a sample of pregnant Black persons seeking prenatal care from a large safety net hospital in an urban community and (b) examine the relations among various types of childhood maltreatment and later IPV with and without a weapon. Specifically, we hypothesized that (H1) there will be high rates of childhood maltreatment and IPV victimization, and (H2) various childhood maltreatment types, especially CSA and CPA, will be positively associated with later IPV victimization (with and without a weapon).
Method
Participants and Procedure
Participants were recruited from the obstetrics clinic of a public hospital in Atlanta, Georgia (n = 186; mean age = 27.2 years, SD = 5.3) for involvement in an ongoing study assessing the impact of maternal trauma exposure and responses on perinatal and obstetric outcomes within an urban population. Notably, the obstetrics clinic was housed in a publicly funded, safety net hospital, meaning the hospital provides medical care regardless of patients’ insurance status or ability to pay for services. In other words, many participants had limited access to medical and economic resources: most were insured via Medicaid (71.5%, n = 133), and many were unemployed (51.1%, n = 95) and had a monthly household income of less than $2,000 (49.0%, n = 91). Trained research assistants contacted participants to determine interest in and eligibility for study participation. Participants were verbally screened for trauma exposure and related outcomes via self-report questionnaires. On average, participants took 45 to 75 min to complete the screening assessment. After completion, participants were thanked for their participation, compensated $40 via cash or an electronic gift card, and debriefed.
For inclusion in the study, participants must have identified as Black or African American, spoken English, initiated prenatal care at Grady Memorial Hospital, been aged 18 to 40, and been within the 10 to 40 weeks gestational period, as determined by a positive urinalysis and fetal monitoring. Exclusion criteria included active psychosis, acute intoxication, admission to a psychiatric hospital within the last month, and any signs of intellectual disability (i.e., inability to articulate an overall understanding of the study’s purpose and procedures). Only participants who responded to one or more items regarding their IPV history were included in the current study. The Emory University Institutional Review Board and Grady Research Oversight Committee approved all study procedures in accordance with the Declaration of Helsinki.
Measures
Demographics
Participants self-reported age, education (less than 12th grade, 12th grade/high school graduate, GED, some college or technical school, technical school graduate, college graduate, or graduate school), health insurance status (no insurance, Medicaid, private insurance), monthly household income (<$250, $250–$499, $500–$999, $1,000–$1,999, ≥$2,000), and current employment status (Employed/Unemployed).
Childhood Maltreatment
The Childhood Trauma Questionnaire (CTQ)—Short Form, a 28-item retrospective self-report measure (Bernstein et al., 1994, 2003), was used to assess the presence and severity of participants’ childhood maltreatment across five subscales: physical abuse (CPA), sexual abuse (CSA), emotional abuse (CEA), physical neglect (CPN), emotional neglect (CEN). Item responses ranged from Never true (1) to Always true (5). The CTQ demonstrates good reliability and validity in clinical and community populations (Bernstein et al., 1994, 2003), including low-income adults seeking public health care (Binder et al., 2008). The overall CTQ score ranges from 25 to 125, and each subtype score ranges from 5 to 25. The CTQ demonstrated excellent internal consistency in this sample (α = .945).
IPV With and Without a Weapon
The Traumatic Events Inventory (TEI; Gillespie et al., 2009), a 14-item self-report measure assessing frequency and type of trauma(s) experienced, was used to assess participants’ exposure to various types of trauma across the lifespan (e.g., experiencing IPV, experiencing childhood abuse [i.e., physical, sexual, emotional]). Construct validity of this measure for this sample has previously been shown (Mekawi et al., 2021). Participants were asked two questions to assess whether they had experienced IPV victimization with/without a weapon (e.g., with a weapon: “Have you been attacked with a gun, knife, or other weapons by a spouse, romantic partner/boyfriend, or girlfriend”; without a weapon: ““Have you been attacked without a gun, knife, or other weapons by a spouse, romantic partner/boyfriend, or girlfriend”; 0 = No, 1 = Yes); if endorsed, the frequency of exposure (One time, Two or three times, Four or five times, Six to eight times, Nine to 10 times, 11 to 15 times, 16 to 19 times, Over 20 times) and age of first occurrence were also assessed. TEI items assessing age of first childhood maltreatment (e.g., childhood physical, sexual, and emotional abuse) were also checked to ensure first child maltreatment experience preceded first IPV experience.
Data Analysis
First, we assessed the prevalence of childhood maltreatment and IPV using data collected from the CTQ and TEI, respectively. If both childhood maltreatment and IPV were endorsed, the ages at which first childhood physical, sexual, and emotional abuse, and first IPV victimization occurred were examined to ensure the first IPV experience, even if in childhood, occurred after the first childhood maltreatment experience. Three participants were removed from further analyses due to our inability to confirm one or more forms of childhood maltreatment occurred before their first IPV victimization (as determined by the TEI). Specifically, these three participants reported their first IPV victimization occurred before age 18, but TEI data regarding their age of first childhood physical, sexual, and/or emotional abuse were missing. Roughly 1.1% (n = 2) of CTQ data were missing. Thus, sample sizes differ across analyses.
Next, the bivariate relations between the five subtypes of childhood maltreatment and IPV were assessed using Mann–Whitney U Tests, as we determined that the CTQ subtype scores were not normally distributed with Shapiro-Wilk tests. We used a Bonferroni correction to adjust the level of significance for multiple comparisons (p = .01 [.05/5 childhood maltreatment types]). For any group comparisons that survived correction for multiple comparisons, the relation between childhood maltreatment subtypes and both IPV types were further assessed using two binary logistic regression models. Model 1 examined the relationship between childhood maltreatment and IPV with a weapon. The binary outcome variable was IPV with a weapon (no or yes), and the predictors were childhood maltreatment subtype scores. Model 2 examined the association between childhood maltreatment and IPV without a weapon. The binary outcome variable was IPV without a weapon (no or yes), and the predictors were childhood maltreatment subtype scores. We checked for outliers by calculating Cook’s distance for each observation. Any observations with a Cook’s distance > 0.5 would be considered outliers; no observations in either model met this threshold. We assessed the possibility of multicollinearity by calculating the variance inflation factors (VIF) for each model; none of the VIFs for either model were > 5. Statistical significance was set at p < .05 for the logistic regression analyses. SPSS (version 29) and R (version 4.1.2) were used for statistical analyses, and pairwise deletion was used.
Results
In line with (H1), many participants endorsed CEA (53.0%, n = 96), CEN (47.5%, n = 86), and CSA (40.6%, n = 73). Roughly 28.7% reported CPN (n = 52), and 39.8% reported CPA (n = 72). A sizable proportion of the sample also reported experiencing IPV (42.6%, n = 78); 38.8% endorsed experiencing IPV without a weapon (n = 71), 21.3% endorsed experiencing IPV with a weapon (n = 39), and 17.5% endorsed experiencing both (n = 32). See Table 1 for sample characteristics and Table 2 for descriptive statistics.
Sample Characteristics, n = 186.
Descriptive Statistics and Proportion of Sample Endorsing Trauma Exposure by Type.
Note. n = 180–183. CTQ = Childhood Trauma Questionnaire; TEI = Traumatic Events Inventory; CSA = childhood sexual abuse; CPA = childhood physical abuse; CEA = childhood emotional abuse; CEN = childhood emotional neglect; CPN = childhood physical neglect; IPV = intimate partner violence.
Proportion of individuals with CTQ subscale scores in at least the low to moderate range.
CTQ subscale cut-off values are presented as “none/minimal” / “low to moderate”/ “moderate to severe” / “severe to extreme” according to Bernstein et al. (2003).
n(%) column refers to n(%) of participants who endorsed IPV with and without a weapon.
Following correction for multiple comparisons, childhood maltreatment differed between those who had experienced IPV with a weapon and those who did not for all childhood maltreatment subtypes except CPN (Table 3). As such, we conducted a follow-up regression analysis to examine associations between exposure to IPV with a weapon and the different childhood maltreatment subtypes that were significant at the bivariate level, which were included as predictors (i.e., CSA, CPA, CEA, and CEN). The logistic regression model (Model 1) revealed that CSA was the only significant predictor of IPV with a weapon (B = 0.10, p = .003; Table 4), suggesting our second hypothesis was partially supported. Holding the other forms of childhood maltreatment constant, the odds of experiencing IPV with a weapon increased by 10% (OR = 1.10, 95%CI [1.04, 1.18]) for every point increase in CTQ CSA subtype score.
Bivariate Relationships Between Childhood Maltreatment and IPV, Assessed With Mann–Whitney U Tests.
Note. Bolded p values indicate significance that remained after Bonferroni correction for multiple comparisons; CSA = childhood sexual abuse, CPA = childhood physical abuse, CEA = childhood emotional abuse, CEN = childhood emotional neglect, CPN = childhood physical neglect, IPV = intimate partner violence.
Binary Logistic Regression Models Examining Associations Among Childhood Maltreatment Subtypes and IPV With and Without a Weapon.
Note. n = 186. CSA = childhood sexual abuse, CPA = childhood physical abuse, CEA = childhood emotional abuse, CEN = childhood emotional neglect, CPN = childhood physical neglect, IPV = intimate partner violence.
Similarly, all childhood maltreatment subtype scores except CPN were significantly different between the participants who reported a history of IPV without a weapon compared to the participants who denied a history of IPV without a weapon, following correction for multiple comparisons (Table 4). We conducted a follow-up logistic regression analysis to examine the associations of IPV without a weapon with types of childhood maltreatment that were significant at the bivariate level entered as predictors (i.e., CSA, CPA, CEA, and CEN). Again, in line with our second hypothesis, CSA was the only significant predictor (B = 0.11, p = .001; Table 4). The odds of experiencing IPV without a weapon increased by 12% (OR = 1.12, 95%CI [1.05, 1.20]) for every point increase in CTQ CSA subtype score.
Discussion
The overall goal of the current study was to characterize rates of childhood maltreatment and IPV victimization and examine the associations between childhood trauma and IPV in a sample of pregnant Black persons seeking perinatal care at an urban safety net hospital. Corroborating previous work (Powers et al., 2020), there were high rates of childhood maltreatment and IPV victimization in the current sample of pregnant Black persons seeking health care services. In this study, most participants endorsed childhood maltreatment, with CEA and CEN being the most frequently reported forms. Schreier et al. (2015) found that among pregnant women, race/ethnicity was significantly associated with childhood trauma, with Black women more likely to report any form of childhood trauma than White or Hispanic women. Notably, in the current sample, the rate of childhood trauma exposure was even higher than those typically reported by Black pregnant persons (Caceres et al., 2022), perhaps due to their intersecting vulnerabilities. Similarly high trauma exposure rates have been observed within other samples of under-resourced, minoritized, health care-seeking populations (Gluck et al., 2021; Powers et al., 2020).
Notably, nearly half (42.6%) of participants in the present studies endorsed a history of IPV victimization. In a previous studies of lifetime trauma exposure, predominantly Black patients were approached in waiting rooms of medical clinics at the same public hospital in which the current study was conducted; 14.4%-18.0% of participants had ever been attacked by a partner with a weapon, and 32.8%-37.5% had ever been attacked by an intimate partner without a weapon (Gluck et al., 2021; Powers et al., 2020). While the percentage of participants reporting a history of IPV in the present study appears consistent with the national lifetime prevalence of IPV among Black women in the United States (43.7%; Black et al., 2011), it should be stated that the mean age of current study participants was only 27 years. Taken together, findings underscore the critical need to screen for IPV during routine healthcare encounters, particularly when multiply marginalized patients are pregnant and/or of reproductive age (Langhinrichsen-Rohling et al., 2021).
In addition, childhood maltreatment, specifically CSA, was associated with later IPV victimization with and without a weapon, independent of other childhood maltreatment exposure. This is consistent with prior research demonstrating strong associations between CSA and revictimization, especially for types of crime involving interpersonal conflict (e.g., sexual and physical assault, stalking; Papalia et al., 2021). Childhood maltreatment, particularly sexual abuse, may increase the risk of later IPV victimization through subsequent emotional dysregulation (Crow et al., 2014; Widom & White, 1997)and greater risk of PTSD, depression, and anxiety symptomatology (Crow et al., 2014; Fergusson et al., 2008) and substance use (Widom & White, 1997). These factors, individually or in combination with other socioeconomic vulnerabilities, may increase risk for exposure to violence more generally and by intimate partners. However, even when controlling for a range of confounding factors, Jennings et al. (2015) found that experiencing CSA was associated with greater odds of experiencing IPV. Although derived from a sample that greatly differs from the present study (i.e., a large national sample of college students), Jennings et al.’s (2015) findings underscore the strength of CSA as a risk factor for IPV victimization.
A key novel aspect of this study is its examination of the associations between childhood maltreatment and IPV, both with and without a weapon. We found that CSA was associated with greater odds of experiencing IPV both with and without a weapon. These findings are notable in that characteristics of IPV and outcomes differ by weapon use (Sorenson, 2017; Zeoli et al., 2016). For example, in a retrospective review of all police calls related to IPV in Philadelphia in 2013, perpetrators who used their body as a weapon or used non-gun external weapons (e.g., knives) were more likely to engage in violent, aggressive behaviors like punching and pushing, which resulted in a greater chance of visible injuries for victims (Sorenson, 2017). Victims of IPV that involved guns were more likely to report being threatened and feeling frightened and less likely to have visible injuries—an important finding to incorporate into education for providers and other community members (Sorenson, 2017). However, perpetrator firearm access has been linked to increased IPV severity, with more severe threats of violence and more severe actual violence—both physical and nonphysical (McFarlane et al., 1998) reported by pregnant individuals with a history of IPV. A systematic review has found some promising, albeit limited, evidence to suggest that interventions that specifically target firearm access in IPV cases reduce the risk of both nonfatal and fatal violence (Zeoli et al., 2016).
Because pregnant people have more regular contact with healthcare providers through their prenatal visits, screening for trauma histories during pregnancy may provide a critical opportunity to identify individuals at increased risk for IPV with or without a weapon (Langhinrichsen-Rohling et al., 2021). Capitalizing on more routine contact with the healthcare system as a result of pregnancy may be especially important for patients with limited access to financial resources and medical care, such as many in the current sample. According to the American College of Obstetricians and Gynecologists (2012), IPV screening and counseling should be conducted periodically throughout obstetrics care, including at the first prenatal visit, at least once per trimester, and at the postpartum check-up; notably, routine or universal IPV screening is associated with higher IPV detection (Ahmad et al., 2017). Based on the limited state of the current literature, however, the US Preventive Services Task Force has hesitated to put forth recommendations regarding childhood maltreatment screening in primary care (Hoft & Haddad, 2017; Viswanathan et al., 2018). However, our results suggest that screening for both CSA and IPV during prenatal visits may help identify individuals at risk for adverse maternal-fetal health outcomes and prompt the creation of a safety plan and the provision of psychoeducation and resources (e.g., local shelters for victims of IPV, accessible PTSD treatment options). Further, interventions that target the prevention and reduction of childhood abuse appear promising (Fortson et al., 2016), such as school-based programs that teach children about their right to safety and protection and how to ask for help, policy changes that provide additional economic support for families (e.g., tax credits) and promote access to affordable, high-quality childcare, and educational campaigns that increase parents’ knowledge of abuse recognition and prevention.
The results of the present study need to be interpreted within the context of its limitations. First, unlike other studies that utilize medical or legal records (e.g., Papalia et al., 2021), participants’ trauma histories were ascertained from self-reported, retrospective instruments (CTQ and TEI) and are thus subject to recall bias. Of note, the TEI is a site-specific, self-report measure developed and used by the clinical research project from which the current study’s data originate, so the measure has not undergone rigorous psychometric evaluation. Second, we did not specifically assess whether IPV had occurred during pregnancy. Third, the focus of trauma during adulthood was limited to IPV, therefore the results may not generalize to other forms of trauma exposure. Finally, the current results may not generalize to the general population, as participants were pregnant Black or African American patients seeking obstetrics care at a public, safety net hospital in Atlanta, Georgia.
In conclusion, childhood maltreatment was associated with later IPV in pregnant Black/African American persons. Consistent with prior work, CSA was associated with later IPV victimization, both with and without a weapon, in a sample of Black individuals seeking prenatal care. Additional research is necessary to understand the factors that contribute to the association between childhood maltreatment and later IPV victimization and to develop effective screening tools and interventions informed by cu findings. These results also underscore the importance of enacting early, community-level initiatives to prevent and address childhood maltreatment, specifically CSA, and to minimize risk of later IPV victimization.
Footnotes
Correction (May 2024):
Funding information has been updated in the paper.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interests with respect to the authorship and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research and/or authorship of this article: This work was supported by the National Institute of Mental Health (MH115174 to VM) and F32MH134528 to MH.
