Abstract
Intimate partner violence (IPV), childhood abuse, and sexual assault in adulthood are profound public health concerns, particularly for women. Exposure to trauma can contribute to long-standing health problems and escalated medical costs. Unfortunately, these experiences are often intertwined. Sexual assault often occurs in intimate relationships in which there is concurrent IPV; likewise, many victims of IPV have experienced childhood abuse. The prevalent intersections of these struggles can lead to posttraumatic stress disorder (PTSD) symptoms. This article examines the contributions of childhood abuse histories and sexual assault to PTSD symptoms among women experiencing IPV. Findings suggest childhood abuse experiences account for more variance in PTSD symptoms than adult sexual assault. Clinical implications are discussed.
Introduction
Posttraumatic stress disorder (PTSD) affects thousands of Americans each year and occurs as a response to a threatening and traumatic experience (American Psychiatric Association [APA], 2000). PTSD is more prevalent in women than men, which may reflect differences in the type of trauma exposure; women are more likely than men to be a victim of physical violence at the hands of an intimate partner, or a victim of rape by a loved one, friend, or stranger (Kessler, Berglund et al., 2005). Physical and sexual intimate partner violence (IPV) and rape outside of a relationship are potent causes of PTSD in women.
Annually, more than 1.3 million American women are victims of IPV, and 17.4% of women experience a rape in their lifetime (Black et al., 2011; Tjaden & Thoennes, 2000). Unfortunately, women who experience victimization in adulthood often report one or more forms of abuse as a child. Hence, many women’s lives are characterized by a continuous cycle of abuse, resulting in profound mental health consequences. The current literature indicates that childhood trauma, whether sexual or physical, can increase the risk of adult sexual and physical victimization (although not necessarily in the context of IPV), leading to intensified PTSD symptomatology (Becker, Stuewig, & McCloskey, 2010; Fogarty, Fredman, Heeren, & Liebschutz, 2008; Nishith, Mechanic, & Resick, 2000; Noll, Horowitz, Bonanno, Trickett, & Putnam, 2003; Schaaf & McCanne, 1998). This article examines the contribution of childhood abuse and adult sexual assault toward vulnerability to PTSD, among a sample of women exposed to IPV.
Background and Significance
PTSD
In the Diagnostic and Statistical Manual of Mental Disorders (5th ed., DSM-5; APA, 2013), PTSD is defined as a trauma-related disorder that occurs in response to a life-threatening event (real or perceived) due to an inability to assimilate the traumatic experience as a memory of the past. PTSD manifests as reexperiencing the event in painful dreams and thoughts, strict avoidance of stimuli associated with the original experience, and a constant state of hyperarousal (APA, 2013). Notably, in the revision of DSM-5, sexual violation is included as an explicit potential trigger for PTSD. Approximately 6.8% of Americans will suffer from PTSD at some point during their lifetime, indicating the devastating impact of this disease on public health (Kessler, Berglund et al., 2005).
Women are at increased risk for developing PTSD, as the lifetime prevalence of a PTSD diagnosis has been estimated as 10.4% in women versus 5.0% in men (Kessler, Chiu, Demler, & Walters, 2005). Women with PTSD are also more likely to be afflicted with a comorbid diagnosis of major depressive disorder (MDD), bulimia nervosa, or a second anxiety disorder, and may experience a greater disease burden than men (McLean, Asnaani, Litz, & Hofmann, 2011). While it was originally thought that higher prevalence of PTSD in women was due to an intrinsic vulnerability (Breslau, Chilcoat, Kessler, Peterson, & Lucia, 1999), the observed gender difference in PTSD prevalence and severity can also be accounted for by the increased incidence of a lifetime history of IPV and rape in women (Cortina & Kubiak, 2006). An investigation of both forms of victimization, beginning with IPV, is necessary to understanding the catastrophic effects that intimate violence can have on women’s mental health.
IPV
IPV entails physical, sexual, or psychological harm by a current or former intimate partner, and is a public health issue that affects millions of people every year. Similar to PTSD, IPV affects more women than men, as reported in community (Tjaden & Thoennes, 2000) and health care clinic (Bonomi et al., 2009; Fishman, Bonomi, Anderson, Reid, & Rivara, 2010) samples. While IPV is associated with a host of negative physical consequences that extend beyond the initial injuries, including gastrointestinal issues, increased pain, and somatic complaints, IPV can also cause a myriad of mental health issues (Campbell et al., 2002). PTSD, depression, anxiety, and suicidal ideation are just some of the mental health consequences of IPV. These insidious physical and psychological health effects are so wide-ranging because IPV encompasses an array of violent physical, mental, and sexual experiences.
One factor within a violent relationship that has been shown to predict the severity of the victim’s PTSD symptomatology as well as physical health status is her real-time fear and perception of danger (Straus et al., 2009). Interestingly, it has been suggested that fear of the abuser and feelings of being unsafe may correlate better with the prevalence of PTSD in the victim than the experience of violence itself (Wilson et al., 2011). Women who rate greater danger in their relationship are more likely to experience anxiety, depression, suicide attempts, and symptoms of PTSD, including difficulty concentrating and memory loss, than those with a lesser sense of danger. These same women are more likely to use tobacco and illicit drugs than their counterparts who feel safer (Peterson, 2013; Sato-DiLorenzo & Sharps, 2007). Thus, the damage caused by IPV cannot be simply quantified by recording the types or frequency of aggression that occurs but also requires a victim-centered exploration of her own interpretation and subjective experience of the abuse.
Sexual Assault
Although the literature on the subject is sparse, IPV is commonly accompanied by sexual assault, thus compounding two independent causes of PTSD into a single abusive relationship. Sexual assault within intimate relationships has been shown to be an independent factor in increasing PTSD symptomatology, beyond the impact of physical abuse (Bennice, Resick, Mechanic, & Astin, 2003). Despite the commonly held belief that rape is most traumatic when committed by an unknown perpetrator, sexual assault in marriage or dating relationships has been found to be equally detrimental to women’s physical and mental health (Kilpatrick et al., 2003). In fact, rape by an intimate partner may increase psychological distress as living with a perpetrator could result in continuous levels of stress and increased dissociation (Temple, Weston, Rodriguez, & Marshall, 2007). In addition, IPV-affected women who experience sexual assault tend to experience more severe concurrent physical violence (Frieze, 1983). Notably, the severity of violence exposure correlates with the severity of PTSD symptomatology (Woods, 2005).
In addition to PTSD, female victims of rape are at substantially increased risk for developing MDD, substance abuse disorders, problems with social adjustment and interpersonal interactions, and anxiety disorders, such as phobias, panic attacks, and obsessive-compulsive disorder (Burnam et al., 1988; Davidson, Tupler, Wilson, & Connor, 1998; Dennis et al., 2009; Jenkins, Langlais, Delis, & Cohen, 1998; Kilpatrick et al., 2003; Resick, Calhoun, Atkeson, & Ellis, 1981). Postrape PTSD can also act as a mediator of physical health symptomatology (Campbell, Ahrens, Sefl, Wasco, & Barnes, 2001; Clum, Calhoun, & Kimerling, 2000; Clum, Nishith, & Resick, 2001; Zoellner, Goodwin, & Foa, 2000). Unfortunately, this complex pattern of psychopathology is often not precipitated by one independent event but can be the result of multiple forms of violence, whether inflicted by a single perpetrator in adulthood or many abusers throughout a lifetime.
Childhood Abuse
Women who experience IPV and sexual assault in adulthood are more likely to have a history of childhood trauma than those who do not; childhood physical and sexual abuse independently increase the likelihood of revictimization in the future and work synergistically in combination (Coid et al., 2001; Desai, Arias, Thompson, & Basile, 2002; Gladstone et al., 2004; Hetzel & McCanne, 2005; Messman-Moore, Ward, & Brown, 2009; Rich, Gidycz, Warkentin, Loh, & Weiland, 2005; Schaaf & McCanne, 1998; Ullman, Najdowski, & Filipas, 2009). Literature shows that women who experience sexual assault before the age of 18 are at greatly increased risk for developing PTSD, MDD, substance abuse problems, eating disorders, and suicidal ideation (Ackard & Neumark-Sztainer, 2002; Fergusson, Boden, & Horwood, 2008; Finkelhor, Ormrod, & Turner, 2007; Gladstone et al., 2004; Kilpatrick et al., 2003; Levitan et al., 1998; Mullen, Martin, Anderson, & Romans, 1993; Rowan, Foy, Rodriguez, & Ryan, 1994; Saunders, 1999; Silverman, Raj, Mucci, & Hathaway, 2001). Childhood sexual assault rarely occurs without being accompanied by other forms of abuse, such as physical abuse, emotional abuse, and neglect; in fact, a single form of abuse takes place in less than 10% of cases (Finkelhor et al., 2007; McGee, Wolfe, Yuen, & Wilson, 1995; Ney, Fung, & Wickett, 1994). Therefore, adults who were sexually assaulted as children, were likely abused in multiple ways and would be at greater, compounded risk of suffering adverse psychological effects from their traumas. Dissociative symptoms, alexithymia, affect dysregulation, depression, and drug and/or alcohol use or abuse may also develop as maladaptive coping reactions to trauma (Meyerson, Long, Miranda, & Marx, 2002; Mulder, Beautrais, Joyce, & Fergusson, 1998). Unfortunately, these coping mechanisms may increase vulnerability to future assault, both physical and sexual, by preventing the victim from recognizing high-risk situations and impeding consequent responses to danger (Classen, Field, Koopman, NevillManning, & Spiegel, 2001; Cloitre & Koenen, 2001; Messman-Moore et al., 2009). In addition, the depressive symptoms that are often comorbid with PTSD may lead to increased financial dependence and decreased motivation to escape an abusive situation (Kim & Capaldi, 2004; Lehrer, Buka, Gortmaker, & Shrier, 2006). Thus, childhood trauma can result in enduring mental health deficits that impede victims from preventing future exposure to violence or sexual assault.
Clearly, sexual and physical abuse in childhood and adulthood are correlated and have profound consequences for women’s mental health. However, research on the cumulative effects of violent and sexual assault over the course of a lifetime is limited. Nishith et al. (2000) found that increased childhood sexual abuse correlated with higher rates of physical and sexual victimization in adulthood, which, in turn, caused increased severity of PTSD after a recent rape not within the context of a physically violent relationship. While this suggests that childhood trauma leads to adult victimization in some form, enhancing dissociative symptoms, several questions remain that require further consideration. As far as we know, an investigation of the impact of violent and sexual assault histories and victims’ perceptions of danger on PTSD symptomatology has not previously been undertaken. Thus, the following examination has important implications for the development of clinical approaches to treatment and prevention of PTSD tailored for the women for whom such a history of victimization is a reality.
We hypothesized that women who had experienced sexual victimization in their intimate relationship, controlling for childhood trauma, would be more likely to experience PTSD. In addition, we posited that a woman’s perception of her danger in these situations would play a central role in determining her PTSD symptomatology.
Method
Secondary data analyses were conducted on a sample of women petitioning for a protection order against an intimate partner, within an upstate New York Family Court. This site offered a relatively diverse demographic cross section. At the time of the study, eligibility for Family Court required the following: being married or divorced, having a child in common, or being related by blood. Dating couples and same-sex couples, even if cohabitating, were not eligible to petition in family court, and were not available for recruitment. Fifty percent of women approached (190 out of 380), consented to take part in the study, and there were no differences in race and age for those that declined. Researchers approached the participants in a private waiting room, out of the presence of perpetrators and court personnel. Study participants were required to be older than the age of 18 years, able to read and write in English, and not too injured to participate. Exclusion criteria included being male or under the influence of drugs or alcohol. Women who provided informed consent completed a series of questionnaires, including measures of PTSD, childhood trauma, and violence. An Institutional Review Board approved all aspects of this study.
Measures
Women (N = 162) completed a demographic survey to assess their age, gender, race, zip code, number of people in household, occupation, household income, and health insurance status and a series of measures. Because the recruitment occurred in a busy court waiting room, some women who provided informed consent needed to leave the area before completing the instruments.
PTSD
PTSD severity was assessed using the Modified PTSD Symptom Scale (PSS), a 17-item self-report measure of PTSD symptomatology frequency and severity over the previous 2 weeks as per the Diagnostic and Statistical Manual of Mental Disorders (3rd ed.; DSM-III; APA, 1980) criteria for PTSD (Foa, Riggs, Dancu, & Rothbaum, 1993). Each symptom was rated on a 4-point scale of frequency and a 5-point scale of severity. Given concerns that we may be overestimating the prevalence of PTSD in this court-based population due to the timing of their interview, we thought a continuous approach to analyzing PTSD symptoms may inflate associations within our data set and opted to stay with a conservative approach of using a cutoff score with the PSS. This cutoff score still only identified 60% of our sample as being above the threshold for PTSD, despite a highly traumatic current lifetime event. Scores of greater than or equal to 34 out of a possible 136 were coded as a PTSD diagnosis (1 = yes, 0 = no) for analytic purposes.
Trauma Exposure
The Traumatic Life Events Scale (TLES; Kubany et al., 2000), a 24-item survey, was used to identify potentially traumatic life events, such as the following: witnessed violence as a child, childhood physical punishment resulting in injury, and unwanted sexual contact prior to and after turning the age of 18 years. Given the interest in examining the potential additive consequences of sexual trauma, responses on the TLES were used to dichotomize the cohort into those who had experienced childhood sexual abuse versus those who had not. For analytic purposes, childhood sexual abuse was defined as present if women endorsed any sexual abuse prior to the age of 18 years on the TLES (1 = yes, 0 = no), Questions 15 and 16. Adult sexual victimization was coded from the TLES Question 18 (1 = yes, 0 = no)
Danger Perceptions
In an effort to assess their perceptions of danger within their relationships, the study subjects also completed the Danger Assessment measure (Campbell et al., 2003) by providing dichotomous answers to questions about known risk factors for IPV, such as past threats to harm the victim’s children, violence during pregnancy, or abuser unemployment or drug use status.
Analysis
Analyses began with an examination of bivariate associations between variables. Next, a path analysis was run using MPlus Version 7 (Muthén & Muthén, 1998-2010), using the weighted least squares with mean and variance (WLSMV) estimator to allow for nonnormality of data. Within the path model, age, race, and poverty were entered as covariates; the presence of a PTSD diagnosis was predicted by childhood and adult sexual abuse, and dangerousness. Adult sexual abuse was predicted by childhood sexual abuse. Model fit was evaluated using global fit indices, including root mean square error of approximation (RMSEA) < .05, comparative fit index (CFI) > .93, and Tucker–Lewis index (TLI) > .73 (Hu & Bentler, 1999).
Results
The effect of each of the independent study variables—including age, poverty level, race, danger category, and history of forced sex and forced sex with use of a threat or weapon—on the dependent variable of PTSD is tabulated in Table 1. Overall, PTSD symptoms were strikingly prevalent in this cohort of women seeking protection orders at family court. A total of 103 of the 162 subjects (64%) had a qualifying score on the Modified PTSD Symptom Scale; age, poverty level, and race did not significantly correlate with PTSD symptoms among these women. Women with significant PTSD symptoms were more likely, than those with minimal symptomatology, to perceive higher levels of danger within their relationship (p < .05), to have been made to have sex against their will (p < .01) by use of physical force or the threat of a weapon (p < .05), and to have experienced comorbid depression (p < .001).
Sample Characteristics by PTSD Endorsement (Bivariate Analysis).
Note. Percentages will not add up to 100% due to missing data. PTSD = posttraumatic stress disorder; TLES = Traumatic Life Events Scale.
A path analysis to determine the presence and significance of hypothesized relationships between independent variables was performed and is tabulated in Table 2 and illustrated in Figure 1. We did not find a significant relationship between childhood sexual abuse and adult sexual victimization in this particular group of women, as illustrated in the first path from the final model in Figure 1. The overarching model was designed to assess the effects of various lifetime risk factors on PTSD symptomatology. Examining this model, we see a direct and statistically significant association between childhood sexual abuse and PTSD symptoms in adulthood. The other direct and statistically significant association seen in this model is between the woman’s assessment of danger in her relationship and current PTSD symptoms. Interestingly, our hypothesis that sexual assault in adulthood by an intimate partner within a milieu of physical violence would independently increase PTSD symptoms was not supported. Although both models were adjusted for age, race, and level of poverty, no sociodemographic variables were associated with PTSD symptoms. Given the diverse group of women surveyed, these data suggest that the woman’s subjective assessment of danger within her relationship trumps her life context.
Model Results.
Note. Race used as a dichotomous variable for analysis. CI = confidence interval.
p < .05. **p < .01.

Path Analysis Illustration.
Discussion
Although the results did not support the hypothesis that women who were sexually assaulted within an intimate relationship would have increased PTSD symptoms, results did indicate that women’s perceived level of danger affected PTSD symptoms. This work has implications on the ways in which women are screened for potential abuse and PTSD symptomatology in mental health and primary care settings, as well as helping agencies such as shelters, legal offices, and victim services. Service providers may believe that because a patient is separated from her partner, she should be safe. However, when patients still perceive danger, even if their perpetrator is incarcerated, these fears must be explored and addressed.
Contrary to previous research, we did not see an association between childhood sexual abuse and adult sexual assault. Similarly, women who were sexually assaulted as adults were no more likely to experience PTSD symptoms than those who were not. Although these results were surprising, it may reflect a unique aspect of this particular sample not yet explored in the literature. For this cohort of domestic violence victims in court, some of whom experienced adult sexual assault, the most relevant predictor of adult PTSD symptomatology was childhood sex assault. This contradicts the idea that the relationship between childhood sex abuse and adult PTSD symptoms is mediated by an increased propensity for adult sexual victimization and suggests a more direct role and conceivably deeper impact of childhood abuse on adult PTSD risk. These findings may also suggest that among this court-based cohort, their adult sexual victimization histories do not account for their PTSD in a statistically significant way given their current levels of danger, thus their visit to court seeking protection.
As expected, perceived danger within the relationship was significant in predicting PTSD symptoms. This suggests that the individual’s experience of the abuse is paramount in molding mental health outcomes and trumps the importance of actual instances of lifetime victimization. Thus, screening for a history of IPV or rape may simply not be enough. Rather, clinicians must understand the victims’ lifetime histories of abuse, whether sexual, physical, verbal, or a conglomerate of the three, and recognize that each woman will view her situation and its inherent risks differently. In addition, a thorough sexual assault history, including childhood abuse, must be taken, despite clinician fears of breaching privacy.
As with any study, there are limitations and findings should be considered with caution. Because the data were obtained via self-report, with no third party verification, there was the possibility of recall bias. The relatively small sample size is also important to consider when extrapolating these data to potential changes to best practice standards. Finally, the study was cross-sectional, thus we were unable to draw causal effects. Although recruitment took place in a busy court setting, a 50% participation rate raises concerns over who refused, especially because there were no discernible differences (in race and age) between those who declined and those who accepted. Because the participants were court-based victims, their PTSD symptoms could have been more severe at the time of recruitment. In addition, these are a subset of women seeking protection orders in civil court who are at a stage of readiness to seek help to stop their violence. More research is needed for how sex assault may affect PTSD symptoms among victims in different settings and different stages of readiness.
Conclusion
Prior to the undertaking of this study, the relationship between childhood trauma, sexual victimization in adulthood concurrent with physical or emotional IPV, and PTSD had yet to be elucidated. The results suggest that PTSD is widespread among a court-based population of victimized women. In addition, these PTSD symptoms are significantly associated with childhood sexual abuse, and an individual woman’s perceived level of danger in their adult intimate relationship, but not with adult sexual abuse. Childhood sexual abuse did not predict adult sex assault in this population, contrary to the literature. These findings suggest that even if IPV victims experience polyvictimization over a lifetime, special attention needs to be paid to how these events shape their day-to-day experience of life; this should prove more valuable than screening in the form of a sterile tally of traumatic life events. In addition, IPV victims with a history of childhood sex abuse may require an integrated treatment regimen with emphasis on their childhood experience as the violence in childhood appears to be a unique and predominant determinant of mental health disturbances in adulthood.
Footnotes
Acknowledgements
The authors wish to thank Jennifer Thompson Stone for her editorial work on this article.
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.
