Predicting Risk Factors for Intimate Partner Violence Among Post-9/11 College Student Veterans

Literature Review

Alarming rates of IPV in marital relationships where one partner is a combat veteran have been well documented across different samples (Bradley, 2007; Morland, Leskin, Block, Campbell, & Friedman, 2008; Solomon, Dekel, & Zerach, 2008; Teten, Sherman, & Han, 2009). A 2005 review (Marshall, Panuzio, & Taft, 2005) of the existing literature on the prevalence of IPV in the military revealed that rates of partner violence ranged from 13.5% to 58%, across military and active veteran samples with lower rates revealed in non-clinical populations. Moreover, self-reported rates of IPV among military couples are higher than rates among civilian couples (Jones, 2012; Stamm, 2009). Heyman and Neidig (1999) found that wives of Army servicemen stationed in at 38 different U.S. installations experienced significantly higher rates of moderate partner violence (13.1% vs. 10.0%) and severe partner violence (4.4% vs. 2.0%) as compared with their civilian counterparts. Moderate violence included having something thrown at them, being pushed, grabbed, shoved, slapped, kicked, bit, or hit with a fist, and severe violence entailed being beaten up, choked, assaulted, or threatened with a weapon. Although Army servicemen themselves reported committing moderate partner violence at rates comparable with those of men in the civilian sample (10.8% vs. 9.9%, respectively), these servicemen reported committing significantly higher rates of severe husband-to-wife violence than did the civilian men (2.5% vs. 0.7%). In another sample of approximately 15% of White male personnel of 38 randomly selected army bases located across the 48 contiguous Unites States, 30% reported perpetrating IPV during the past year (Pan, Neidig, & O’Leary, 1994).

Psychiatric symptoms appear to play a primary role in IPV perpetration. Recent studies of different veteran samples have revealed that veterans with PTSD (Byrne & Riggs, 1996; Marshall et al., 2005; Savarese, Suvak, King, & King, 2001) or depression (Sherman, Sautter, Jackson, Lyons, & Han, 2006) are more likely to perpetrate violence toward their partners as compared with veterans without these disorders. It is estimated that up to 35% (700,000) post-9/11 Global War on Terror veterans experience some form of posttraumatic stress (Institute for Operations Research and the Management Sciences, 2009). In a recent study by Taft et al. (2009) of post-9/11 recent tour veterans, 33% of veterans experiencing posttraumatic stress were identified as domestic violence abusers. More specifically, in a recent study of 129 male combat veterans screened at the National Center for PTSD/VA Boston Healthcare System between 2003 and 2008, Taft and colleagues (2012) found that PTSD was particularly associated with higher rates of IPV particularly in men who had not experienced head contact events (e.g., concussion). Other studies of combat veterans reveal a significant relationship between the severity of PTSD symptoms and the severity of domestic violence perpetrated (Finley, Baker, Pugh, & Peterson, 2010; Gerlock, 2004; Sayers, Farrow, Ross, & Oslin, 2009).

In a study of couples seeking relationship therapy at a Midwestern Veterans Affairs Clinic between 1997 and 2003, Sherman et al. (2006) found that more than 80% of depressed veterans committed at least one act of violence (e.g., slapping partner) in the last year, and more than 40% committed at least one severe act of violence (e.g., choking partner). Studies of both Vietnam era veterans, and of a cross sample of veterans seeking help at a VA clinic, reveal that combat veterans with depression, as well as those with PTSD, display higher levels of anger than do combat veterans without either disorder, supporting the hypothesis that anger, rather than combat experience itself, may best predict the likelihood of perpetrating IPV (Novaco & Chemtob, 2002; Sherman et al., 2006). Among OIF/OEF veterans with current or recently separated partners, 53.7% reported incidents of shouting, pushing, or shoving, and 27.6% reported that their partner was “afraid of them” (Sayers et al., 2009). The same study revealed that depressed veterans identified acts of anger or violence as the reason for seeking counseling.

Ample evidence exists that abusing alcohol and other drugs increases the propensity to perpetrate IPV (Hurlbert, Whittaker, & Munoz, 1991; Kantor & Straus, 1987; O’Keefe, 1997; Straus, 2000). According to the Substance Abuse and Mental Health Services Administration (SAMHSA, Office of Applied Studies, 2007), 7.1% of veterans met the criteria for past year substance use disorder (SUD), and a breakdown by age reveals that 25% of veterans aged 18 to 25 meet those criteria. Veterans with PTSD and depression are highly likely to self-medicate through the use of alcohol and other drugs, with about 75% to 79% meeting the criteria for co-morbid substance abuse disorders (Kulka et al., 1990; L. A. Taylor, Kreutzer, Demm, & Meade, 2003). Marshall et al. (2005) reported that IPV rates among veterans suffering from substance abuse disorders and PTSD range from 1 to 3 times higher than IPV rates in the general population, and the severity of substance abuse is correlated with the frequency and severity of partner abuse.

The structure and culture of military life itself pose considerable challenges for veterans returning home after deployments, because many of the skills and behaviors that are considered adaptive and necessary for survival during war conflict with behaviors that are considered appropriate in the civilian world (Demers, 2008; Hoge, 2010). Military recruits learn to separate themselves from individual emotions, depersonalize acts of violence, and dehumanize the other, to successfully fight wars. Moreover, recruits use battlemind, the “inner strength to face fear and adversity in combat with courage” (Adler, Bleise, McGurk, Hoge, & Castro, 2009, p. 930), in part, through the utilization of aggression. These attitudes are closely related to hypermasculinity, an extreme and rigid adherence to beliefs and characteristics associated with the masculine gender role (Jones, 2012; Katz, 2006; Kivel, 1992; Morris, 1996), and are associated with adherence to domestic violence myths, a set of attitudes and beliefs that serve to justify and excuse committing IPV (Katz, 2006; Kivel, 1992; Peters, 2008; Tolman, 1989, 1999). Morris (1996), for example, proposed that in some military units, men bond with each other around stereotypically masculine characteristics such as dominance, aggression, risk taking, and develop and maintain shared attitudes that support anger toward women and the abuse of female partners. Supporting this suggestion, Rosen, Kaminski, Parmley, Knudson, and Fancher (2003) found that in a sample of married male soldiers who were stationed in Alaska in 1998, increased cohesion and group disrespect, characterized by rude aggressive behavior, conversations that degrade women, consumption of pornography, sexualized discussion, and the encouragement of alcohol use, was associated with increased mean frequency of IPV at both the group and the individual levels. Using cultural spillover theory as a framework, Bradley (2007) analyzed data from the National Survey of Families and Households Wave 1 (1987-1988) to determine whether military training and culture were related to higher rates of IPV among veterans who had separated from military and their spouses. Cultural spillover theory (Baron & Straus, 1987, 1989) suggests that the more a given subculture supports the use of violence to attain socially approved ends, the greater the likelihood that this legitimization of violence will spill over, or be generalized, into other spheres of life. Contrary to expectations, Bradley found that, in his sample, veterans were no more likely than non-veterans to engage in IPV. Furthermore, it is possible that the length of separation time from the military acted as a mediating factor, and that the population under investigation had become sufficiently reintegrated and re-acculturated into civilian norms so as to be protected from IPV risk.

Furthermore, the recent literature suggests that many veterans experience a crisis of identity and report a deep sense of disconnection from family members and other civilians when they return from war and separate from the military (Armstrong, Best, & Domenici, 2005; Demers, 2009; Hoge, 2010). In a meta-analysis of risk factors for PTSD in trauma-exposed adults, Brewin and his colleagues (2000) found that, across the existing literature, factors such as lack of social support and life stress were more strongly linked to trauma effects as compared with individual-level vulnerability. Indeed, social support appears to be inversely related to ongoing effects of PTSD, including relationship problems, as suggested by a 2004 study of Vietnam veterans (Schnurr, Lunney & Sengupta, 2004). A lack of positive social support has also been identified as a primary correlate of IPV, with evidence suggesting that men who feel disenfranchised may use violence against female partners as a way to establish a sense of dominance or social control (Renzetti, 2011).

In sum, the current returning veteran population experiences unique challenges related to reassuming relationship roles and managing hostility. A preponderance of the literature reveals that perpetrators of intimate violence are characterized by reliance on psychological aggression in conflict management, low verbal negotiation skills, adherence to domestic violence myths, and marked anger reactions (Barnett, Miller-Perrin, & Perrin, 2010). Distinct risk factors for perpetrating partner violence in veterans include depression, anxiety, substance abuse, and hypermasculinity. Given the paucity of the literature about IPV risk and co-morbid problems among student veterans on college campuses, this study was undertaken to understand specific factors that predict aggression in the intimate partner relationships of post-9/11 student veterans. It was hypothesized that based on existing research, psychological distress and substance abuse would play a primary role in predicating the characteristics associated with intimate violence perpetration in veterans (Elbogen, Beckham, Butterfield, Swartz, & Swanson, 2008). Based on extant studies, veterans’ identification with hypermasculine attitudes (Jakupcak, Blais, Grossbard, Garcia, & Okiishi, 2013; Katz, 2006) was believed to predict IPV risk beyond the primary effects of psychological symptoms. Conversely, as supported by relevant research, we predicted that social support would reduce the likelihood of aggression, hostile attitudes, and anger in relationships, even in the face of psychological strains and hypermasculinity characteristic of military veterans (Hoge, 2010; Hoge et al., 2004).

Method

Results

The current study provides a descriptive snapshot of IPV in male college student veterans. Although the majority of our sample of college student veterans (80%) indicated satisfaction with their relationship, approximately one third (36%) reported problems in their relationship and 25% endorsed beliefs related to IPV. Veteran students’ beliefs were supported by their self-reported behaviors: 68% insulted or swore at their partner, 67% had shouted at their partner, 12% had threatened to hit or throw something at their partner, 12% had grabbed their partner, 10% had destroyed something of partner’s, and 28% had coerced their partner to have sex with them.

To understand risk factors for IPV, we explored the ways in which psychological distress, substance use, hypermasculine attitudes, and social support were related to psychological aggression, using verbal negotiation tactics instead of violence, domestic violence myth endorsement, and anger. Individuals reporting high levels of psychological distress, substance use, and hypermasculine attitudes were predicted to have greater anger reactions, lower use of negotiation strategies, higher adherence to domestic violence supportive beliefs, and high levels of psychological aggression.

Existing research provided the basis to examine psychological distress and substance abuse as primary predictors of risk for IPV. Based on extant studies, veterans’ identification with hypermasculine attitudes was believed to predict IPV risk beyond the effects of psychological symptoms. In contrast, building on the existing literature, social support was understood as decreasing the likelihood that a veteran would commit IPV, even in the face of mental health concerns, substance abuse, and stereotypically masculine attitudes. Correlational and regression analyses were conducted to examine the relationships between these variables.

Discussion

It appears that anger, relationship dynamics, and IPV are much more complex for veterans than previously reported. Although mean levels of psychological distress, and substance abuse were not significant among the current group, our sample consists of a higher functioning population of veterans than the non-student veteran population usually studied in that they were all enrolled in college, involved in relationships with a significant other, and both willing and able to complete a lengthy survey. Nevertheless, item analysis revealed that approximately 20% to 30% of this group are highly distressed, abusing alcohol and/or drugs, and at risk of partner abuse. In fact, self-reported rates of moderate IPV among this sample were comparable with rates reported by veterans in other studies, such as 12% versus 11% to 13% found by Heyman and Neidig (1999) in their examination of servicemen randomly sampled from 38 army installations between 1990 and 1994.

Correlational analyses revealed significant relationships between certain variables (See Appendix B), suggesting that rates of IPV may be explained by the negative relationship identified between social support and psychological distress as well as the inverse relationship between distress and using verbal negotiation strategies to handle conflict. Again, although the direction of relationship between the experience of distress, and low social support cannot be determined without longitudinal data, both existing research and current findings reveal that aggression is likely to occur when participants experience high levels of psychological distress, anger, and low levels of social support (Lento-Zwolinski, 2007). Unfortunately, in addition to being highly distressed, one third of respondents in this study reported low levels of social support and feeling lonely, and one fourth reported having no close friends. The results of our correlations support a body of literature that suggests that though social support may reduce anger reactions and aggression (Babcock, Roseman, Green, & Ross, 2008; Barrera et al., 2013; Cohen et al., 2000; Csibi & Csibi., 2011; Pietrzak et al., 2009), a lack of social support places veterans at significant risk of using aggressive strategies instead of verbal negotiation to manage conflict (Barrera et al., 2013; Boscarino, 1995; Cohen et al., 2000; Csibi & Csibi, 2011; Pietrzak et al., 2009). Of course, correlations cannot fully explain rates of IPV; they serve only to quantify the relationships among the variables.

In addition to the direct relationships demonstrated between social support, psychological distress, and psychological aggression, in keeping with previous studies, current results demonstrate a strong positive relationship between psychological distress and anger. Furthermore, our findings reveal that distressed and angry veterans are more likely to endorse hypermasculine attitudes and behaviors, such as condoning both physical fighting and risk tasking. As demonstrated by a growing body of literature, hypermasculinity places men at greater risk of using aggressive strategies to manage interpersonal conflict as opposed to using verbal negotiation tactics. Although findings regarding the effects of military training and culture on veterans’ rates of IPV are inconclusive, it is possible that cultural spillover theory can account for hypermasculinity in the current sample. The unique stressors faced by OIF/OEF veterans (prolonged and repeated deployments with little downtime, lack of clear battle lines) may place this group of veterans at special risk of developing hypermasculine attitudes that translate into relationship aggression on the home front. Our findings are in line with current research on more than 2,000 OIF army combat soldiers that suggests that mental health providers should consider focusing on risky behaviors instead of overall posttraumatic stress response (Harmon et al., 2012).

In summary, there is a reciprocal relationship between social support, psychological distress, anger, and the use of verbal negotiation tactics, which has significant implications for veterans and their loved ones. In keeping with a strong body of literature, based on the correlations found, the more social support a veteran experienced, the less distress he experienced and the more likely it was that he used verbal negotiation tactics in his intimate relationship, and a lower likelihood of using aggression.

Regression analyses may shed light on predicting signs of likelihood of aggression in warriors: High levels of psychological distress and hypermasculine attitudes are significant predictors of anger and psychological aggression. Social support seems to negatively predict psychological aggression, reducing the likelihood of this form of abuse even in the presence of other risk factors. This model suggests that IPV cannot be predicted by anger alone (as hypothesized by Sherman et al., 2006) but may be the result of a much more complex constellation of factors.

It may be that military training and culture play a role in hypermasculine attitudes and anger (as a coping mechanism during combat enabling veterans to suppress feelings of fear or sadness that may put the veteran and his or her comrades at risk), and anger may be related to military separation and transition to civilian culture. The relationship between social support, psychological aggression, and hypermasculinity is also complex, with social support appearing to reduce the likelihood of psychological aggression. It is unknown if anger and/or hypermasculine attitudes prevent the development of strong social support, and each of these areas warrants additional exploration.