Abstract
Background
Risk factors of violence perpetration in veterans include substance use and posttraumatic stress disorder (PTSD); however, it is unknown whether these factors are associated with greater risk for partner or non-partner violence. This study investigated the associations between probable PTSD, heavy drinking, marijuana use, cocaine use, and partner and non-partner violence perpetration.
Methods
Self-report questionnaires assessing past-year partner and non-partner aggression (CTS2) as well as past-month substance use (SAOM), probable PTSD (PCL-C), and probable depression (PHQ-9) were administered to 810 substance using veterans entering VA mental health treatment.
Results
In bivariate analyses, probable PTSD in substance using veterans was associated with violence perpetration (partner physical, χ2 = 11.46, p = 0.001, φ = .12; non-partner physical, χ2 = 50.64, p < 0.001, φ = .25; partner injury, χ2 = 6.41, p = 0.011, φ = .09; non-partner injury, χ2 = 42.71, p < 0.001, φ = .23). In multiple logistic regression analyses that adjusted for sociodemographic characteristics, probable PTSD was independently associated with non-partner physical (odds ratio [OR], 2.82; 95% confidence interval [CI], 1.97 – 4.05) and injury aggression (OR, 3.96; CI, 2.56 – 6.13). Cocaine and heavy drinking were independently associated with non-partner physical and injury aggression and non-partner injury aggression respectively.
Conclusions
The results provide evidence that probable PTSD, heavy drinking, and cocaine use are associated with increased risk of non-partner violence perpetration in substance using veterans. These results underscore the importance of screening for PTSD symptoms and violence perpetration towards non-partners in substance using veterans presenting for treatment.
Keywords: Partner aggression, Non-partner aggression, PTSD, Substance use
1. Introduction
Violence perpetration among veterans has emerged as a critical public health concern (MacManus et al., 2015; Sontag and Alvarez, 2008). To date, however, there has been relatively little research investigating factors related to violence among veterans in VA mental health treatment. Prior research has indicated that individuals with substance use and substance use problems have an increased risk of perpetrating violence (Chermack et al., 2010; Chermack et al., 2014; Harford et al., 2013; Pulay et al., 2008) towards intimate partners (Okuda et al., 2015; Singh et al., 2014) and non-partners (Murray et al., 2008). Moreover, past year estimates of violence perpetration in substance use treatment samples are 2–3 times higher than community samples (Chermack et al., 2001). Problematic alcohol and cocaine use have been consistently linked to violence (Chermack et al., 2014; Macdonald et al., 2003), and some research indicates that marijuana use may be more strongly associated with partner violence than non-partner violence (Epstein-Ngo et al., 2014; Moore et al., 2008). Thus, the association between substance use and violence perpetration in veterans is an important area of investigation.
Posttraumatic stress disorder (PTSD) is a common comorbid condition among those with substance use (Najavits, 2005; Najavits et al., 1997), and has also been associated with elevated rates of violence perpetration (Elbogen et al., 2014; MacManus et al., 2015; Marshall et al., 2005; Taft et al., 2009a; Taft et al., 2009b). In the general population, PTSD may be associated with an increased risk of partner violence among men (Hahn et al., 2015), and veterans diagnosed with PTSD have reported significantly higher rates of past-year partner physical aggression compared to veterans without a PTSD diagnosis (Teten et al., 2010). Studies in substance-using samples have also reported that the presence of PTSD increases the likelihood of violence (Crane et al., 2014; Heinz et al., 2015). Given the high comorbidity between PTSD and substance use disorders in veterans (Seal et al., 2011), the associations between substance use, PTSD, and violence are an important focus of research.
The few studies that have examined substance use, PTSD, and violence among veterans have been equivocal. For example, one study reported that comorbid diagnoses of PTSD and alcohol use disorder were linked with greater risk of violence perpetration than either PTSD or alcohol use alone (Elbogen et al., 2014). Similarly, another study found that patients with comorbid PTSD and hazardous substance use reported higher levels of partner physical aggression than patients with only PTSD (Owens et al., 2014). A third study found that substance use and PTSD symptom severity were both related to aggressive behavior (Heinz et al., 2015). Another study found that PTSD symptom severity was associated with physical aggression among veterans, however, alcohol problems were not (Stappenbeck et al., 2014). A more comprehensive analysis is needed to disentangle the associations between PTSD, substance use, and violence perpetration.
A noteworthy limitation of the prior research in veterans is the lack of data on partner versus non-partner violence. Studies in non-veteran populations have reported that different factors may be associated with partner violence compared with non-partner violence (Epstein-Ngo et al., 2014). Specifically, Epstein-Ngo and colleagues (2014) found that heavy drinking and cocaine use were associated with non-partner aggression whereas marijuana was associated with partner aggression. Another limitation in prior veteran-focused work is the lack of research on specific substances other than alcohol. The current study addresses these gaps in the literature by examining both partner and non-partner violence within the same sample of treatment seeking veterans, and investigating whether and to what extent substance use and probable PTSD are related to physical violence and violence perpetration that leads to injury of another person. After adjusting for covariates that have been associated with violence perpetration (Elbogen et al., 2014; Murray et al., 2008; Stappenbeck et al., 2014; Taft et al., 2009a); it was hypothesized that PTSD would be associated with physical violence towards a partner (partner physical aggression), physical violence towards a person other than a partner (non-partner physical aggression), more severe violence resulting in the injury of a partner (partner injury aggression), and violence resulting in the injury of a person other than the perpetrator’s partner (non-partner injury violence). Additionally, based on previous research (Epstein-Ngo et al., 2014), it was hypothesized that past-month heavy alcohol use and cocaine use would be associated with all four types of violence whereas marijuana would only be associated with partner physical and injury aggression.
2. Methods
2.1 Participants and study design
Eight hundred and forty-one veterans completed self-report measures as part of screening for a randomized controlled trial (RCT) of a substance use and violence prevention intervention (#NCT01337973) at a Veterans Affairs (VA) hospital in the Midwest. Due to missing data, 31 veterans were not included in the current analyses leading to a total sample of 810 veterans. Participants were recruited from three different clinics in a VA, namely, substance use disorder, substance use intensive outpatient, and mental health clinics, using posters, presentations at treatment groups, and direct clinician referral. Veterans eligible for the RCT screening reported current or recent substance use or substance use within the past six months and were entering or recently engaged in substance use or mental health treatment. Exclusion criteria for the screen was developed with the RCT intervention study in mind, and included inability to read or speak English, current suicidal ideation, active psychosis, not able to provide informed consent, and having a legal guardian. Further, veterans were excluded if they had insufficient cognitive orientation due to acute substance use, active participation in another intervention study, or they resided outside of the study catchment area. Veterans provided written informed consent prior to completing screening measures. The screen process took approximately 60 minutes and consisted of a variety of self-report measures. Research assistants were present to assist participants if needed, and participants were compensated $10 in gift cards for their time. The VA’s institutional review board approved all procedures.
2.2 Measures
2.2.1. Past-year partner and non-partner violence
Past-year partner and non-partner violence was assessed using the physical aggression (less severe violence) and injury aggression (more severe violence) scales from a modified version of the revised Conflict Tactics Scale (CTS2; Straus et al., 1996). Items on these scales of the CTS2 include items related to physical aggression (“I pushed or shoved my partner”) and aggression that leads to injury (“My partner had a sprain, bruise, or small cut because of a fight with me”). Participants rated the frequency of each item on a 0 “Never” to 6 “20+ times” scale. The measure was modified to assess for non-partner physical and injury aggression by re-administering items with altered wording (“I pushed or shoved someone other than my partner,” “Someone other than my partner had a sprain, bruise, or small cut because of a fight with me; ” see Chermack et al., 2008). The CTS2 has good internal consistency and has been well validated (Simpson and Christensen, 2005). Further, the modified version has been shown to have strong internal consistency (alpha > .90) for measuring both partner and non-partner aggression (Epstein-Ngo et al., 2014). The Cronbach’s alphas for each subscale used in the current sample were: partner physical aggression α = .89, non-partner physical aggression α = .90, partner injury aggression α = .71, and non-partner injury aggression α = .84. For the current study, four dichotomous variables were created from participants’ responses indicating any incidence of aggression on one of the four subscales in the past year.
2.2.2. Probable PTSD
PTSD symptoms were measured through the PTSD Checklist for civilians (PCL-C; Weathers et al., 1994). The PCL-C is a 17 item self-report based on DSM-IV-TR criteria for PTSD. Participants rated (1 “Not at all” to 5 “Extremely”) how bothered they were in the past month for each item – yielding a total severity score between 17 and 85. The Cronbach’s alpha for the PCL-C in the current sample was .95. Participants were identified as having probable PTSD if they endorsed a criterion A event and had a PCL-C score of 50 or higher (Ruggiero et al., 2003). Due to an administrative error, the last item (“Feeling jumpy or easily startled”) of the PCL-C was not administered to participants. To adjust for this, the mean PCL-C item rating was calculated for each participant and added to each person’s total score.
2.2.3. Probable depression
Symptoms of depression were measured by the depression module of the Patient Health Questionnaire (PHQ-9; Kroenke et al., 2001). The PHQ-9 is a 9 item self-report based on DSM-IV-TR criteria for depression. For each of the nine items, participants rated (0 “Not at all” to 3 “Nearly every day”) how bothered they were in the past two weeks. Total scores range from 0 to 27. The Cronbach’s alpha for the PHQ-9 in the current sample was .90. The current study utilized a cut off score of 15 (moderate depressive symptoms) to categorize probable depression (Kroenke et al., 2001).
2.2.4. Heavy alcohol and substance use
Heavy alcohol and substance use were assessed using items from the University of Arkansas Substance Abuse Outcomes Module (SAOM; (Smith et al., 1996). To measure heavy alcohol use, participants were asked to indicate how many days in the 30 days prior to entering substance use treatment they had five or more standard drinks containing alcohol. Marijuana and cocaine were assessed by having participants indicate how many days in the 30 days prior to treatment they used each substance. The SAOM has good test-retest reliability (Smith et al., 1996).
2.3 Data analysis
Statistical analyses were conducted using IBM SPSS 23 for Windows. Separate chi-square tests were used to examine associations between PTSD and each type of aggression (partner physical, non-partner physical, partner injury, non-partner injury). Mann-Whitney U tests were conducted to examine associations between each of the substance use variables and each type of aggression. Multivariable logistic regression modeling was conducted to examine associations between independent variables of interest, covariates, and each type of aggression outcome. Based on prior research, covariates included age, gender, race, income, and probable depression; and the variables of interest were probable PTSD, heavy alcohol use, marijuana use, and cocaine use.
3. Results
3.1. Sample characteristics
Sample demographics are presented in Table 1. Participants’ ages ranged from 21–75 years with a mean of 48.10 years (SD = 13.27). The majority of participants were male (93.3%), not partnered (68.9%) and White (72.5%). Almost half of the sample (45.6%) reported income less than $20,000, 42.1% of participants reported income between $20,000 and $60,000, and 12.3% of participants reported income over $60,000. Approximately one third of the participants had probable depression (31.7%). One hundred eighty-seven participants (23.1%) reported perpetrating partner physical aggression, and 274 participants (33.8%) reported non-partner physical aggression. Seventy-six participants (9.4%) reported perpetrating partner injury aggression, and 148 participants (18.3%) reported non-partner injury aggression in the past year. The majority of participants (82.5%) experienced at least one traumatic event, and 39.1% of the sample was identified as having probable PTSD, based on experiencing a traumatic event and experiencing significant PTSD symptoms. Sixty percent of the sample engaged in heavy drinking, 37.9% used marijuana, and 15.8% used cocaine in the 30 days prior to treatment. Overall, participants averaged 8.99 (SD = 11.28) days having five or more drinks, 4.47 (SD = 9.32) days using marijuana, and 1.72 (SD = 5.58) days using cocaine or crack in the 30 days before entering treatment.
Table 1.
Demographic characteristics of the full cohort (N = 810).
| Characteristics | n | % | Mean (SD) |
|---|---|---|---|
| Age | 810 | 48.10 (13.27) | |
| Partner status | |||
| Partnered | 252 | 31.1 | |
| Not Partnered | 558 | 68.9 | |
| Gender | |||
| Male | 756 | 93.3 | |
| Female | 54 | 6.7 | |
| Race | |||
| White | 587 | 72.5 | |
| Non-White | 223 | 27.5 | |
| Income | |||
| less than 20,000 | 369 | 45.6 | |
| 20,000 – 60,000 | 341 | 42.1 | |
| more than 60,000 | 100 | 12.3 | |
| Probable PTSD | |||
| PTSD | 317 | 39.1 | 61.94 (8.30) |
| No PTSD | 493 | 60.9 | 34.19 (10.70) |
| Probable Depression | |||
| Depression | 257 | 31.7 | 19.31 (3.29) |
| No Depression | 553 | 68.3 | 7.18 (4.21) |
| Heavy drinking | 465 | 57.4 | 8.99 (11.28) |
| Marijuana | 253 | 31.2 | 4.47 (9.32) |
| Cocaine | 125 | 15.4 | 1.72 (5.58) |
| Partner physical aggression | 187 | 23.1 | |
| Non-partner physical aggression | 274 | 33.8 | |
| Partner injury aggression | 76 | 9.4 | |
| Non-partner injury aggression | 148 | 18.3 |
3.2. Bivariate analyses
Chi-square results are presented in Table 2. Unadjusted chi-square analysis found that probable PTSD was associated with higher rates of physical aggression towards partners (χ2 = 11.46, p = 0.001, φ = .12) and non-partners (χ2 = 50.64, p < 0.001, φ = .25). Probable PTSD was also associated with injury aggression towards partners (χ2 = 6.41, p = 0.011, φ = .09) and non-partners (χ2 = 42.71, p < 0.001, φ = .23).
Table 2.
Associations between PTSD and violence perpetration in veterans.
| Partner physical aggression | Non-partner physical aggression | ||||||||||
| PTSD group | Total n | n | % | χ2 | p | φ | n | % | χ2 | p | φ |
| No PTSD | 493 | 94 | 19.1 | 11.46 | 0.001 | 0.12 | 120 | 24.3 | 50.64 | < 0.001 | 0.25 |
| PTSD | 317 | 93 | 29.3 | 154 | 48.6 | ||||||
| Partner injury aggression | Non-partner injury aggression | ||||||||||
| PTSD group | Total n | n | % | χ2 | p | φ | n | % | χ2 | p | φ |
| No PTSD | 493 | 36 | 7.3 | 6.41 | 0.011 | 0.09 | 55 | 11.2 | 42.71 | < 0.001 | 0.23 |
| PTSD | 317 | 40 | 12.6 | 93 | 29.3 | ||||||
Mann-Whitney U analyses demonstrated that heavy drinking was associated with higher rates of non-partner injury aggression (U = 43,351.50, p = .022, r = .08). Marijuana use was associated with higher rates of physical aggression towards partners (U = 53,219.00, p = .029, r = .08) and non-partners (U = 65,258.50, p = .002, r = .11) as well as injury aggression towards both partners (U = 24,715.50, p = .046, r = .07) and non-partners (U = 43,652.00, p = .012, r = .09). Cocaine use was associated with higher rates of physical aggression towards partners (U = 52,144.50, p = .001, r = .12) and non-partners (U = 64,531.00, p < .001, r = .16) as well as injury aggression towards both partners (U = 24,619.00, p = .007, r = .09) and non-partners (U = 43,612.00, p = .001, r = .12).
3.3. Multivariable analyses
Multivariable logistic regression results are presented in Table 3. Collinearity diagnostics indicated no concerns of multicollinearity. Older ages (OR = .98) were associated with reduced odds of partner physical aggression, and being married or living with a partner (OR = 2.01) and non-white race (OR = 2.03) were associated with increased odds of partner physical aggression. Non-significant factors included gender, income, probable PTSD, probable depression, heavy drinking, marijuana use, and cocaine use. Older ages (OR = .96) were associated with reduced odds of non-partner physical aggression. Non-white race (OR = 1.77), probable PTSD (OR = 2.82), and cocaine use (OR = 1.05) were associated with increased odds of non-partner physical aggression. Partner status, gender, income, probable depression, heavy drinking, and marijuana use were non-significant factors.
Table 3.
Multivariate models of violence perpetration in veterans (N = 810).
| Partner physical aggression | Non-partner physical aggression |
Partner injury aggression | Non-partner injury aggression |
|||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Factor | OR | 95% CI | p | OR | 95% CI | p | OR | 95% CI | p | OR | 95% CI | p |
| Age | 0.98 | .96–.99 | 0.001 | 0.96 | .95–.97 | < 0.001 | 0.97 | .95–.99 | 0.002 | 0.96 | .94–.97 | < 0.001 |
| Partner status | ||||||||||||
| Partnered | 2.01 | 1.38–2.92 | < 0.001 | 0.75 | .52–1.08 | 0.121 | 1.76 | 1.04–3.03 | 0.035 | 0.64 | .40–1.03 | 0.065 |
| Not Partnered | REF | REF | REF | REF | ||||||||
| Gender | ||||||||||||
| Male | REF | REF | REF | REF | ||||||||
| Female | 1.32 | .70–2.48 | 0.384 | 0.84 | .45–1.55 | 0.576 | 1.95 | .90–4.20 | 0.089 | 0.27 | .10–.72 | 0.009 |
| Race | ||||||||||||
| White | REF | REF | REF | REF | ||||||||
| Non-White | 2.03 | 1.39–2.97 | < 0.001 | 1.77 | 1.23–2.55 | 0.002 | 1.9 | 1.11–3.25 | 0.019 | 1.73 | 1.12–2.69 | 0.015 |
| Income | ||||||||||||
| less than 20,000 | 1.65 | .92–2.95 | 0.092 | 1.3 | .76–2.21 | 0.334 | 2.79 | 1.03–7.56 | 0.044 | 1.6 | .78–3.19 | 0.208 |
| 20,000 – 60,000 | 1.23 | .69–2.18 | 0.489 | 1.01 | .60–1.72 | 0.962 | 2.18 | .81–5.87 | 0.122 | 1.52 | .76–3.05 | 0.239 |
| more than 60,000 |
REF | REF | REF | REF | ||||||||
| Probable PTSD | ||||||||||||
| PTSD | 1.34 | .91–1.99 | 0.141 | 2.82 | 1.97–4.05 | < 0.001 | 1.19 | .68–2.08 | 0.546 | 3.96 | 2.56–6.13 | < 0.001 |
| No PTSD | REF | REF | REF | REF | ||||||||
| Probable Dep. | ||||||||||||
| Depression | 1.4 | .94–2.08 | 0.101 | 0.9 | .62–1.31 | 0.573 | 1.76 | 1.01–3.07 | 0.047 | 0.56 | .35–.88 | 0.012 |
| No Depression | REF | REF | REF | REF | ||||||||
| Heavy drinking | 1.01 | .99–1.03 | 0.191 | 1.01 | .99–1.02 | 0.348 | 1.02 | .99–1.04 | 0.083 | 1.02 | 1.00–1.04 | 0.017 |
| Marijuana | 1.00 | .98–1.02 | 0.842 | 1.00 | .99–1.02 | 0.643 | 1.00 | .97–1.02 | 0.904 | 1.01 | .99–1.03 | 0.624 |
| Cocaine | 1.02 | .99–1.05 | 0.121 | 1.05 | 1.02–1.09 | < 0.001 | 1.02 | .98–1.05 | 0.430 | 1.03 | 1.00–1.06 | 0.048 |
Older ages (OR = .97) were associated with reduced odds of partner injury aggression. Being married or living with a partner (OR = 1.76), non-white race (OR = 1.90), income less than $20,000 (OR = 2.79), and probable depression (OR = 1.76) were associated with increased odds of partner injury aggression. Gender, probable PTSD, heavy drinking, marijuana use, and cocaine use were non-significant factors in the model. Older ages (OR = .96), being female (OR = .27), and probable depression (OR = .56) were associated with reduced odds of non-partner injury aggression. Non-white race (OR = 1.73), probable PTSD (OR = 3.96), heavy drinking (OR = 1.02), and cocaine use (OR = 1.03) were associated with increased odds of non-partner injury aggression. Non-significant factors were partner status, income, and marijuana use.
4. Discussion
The current study investigated the association between probable PTSD and past-year violence perpetration in 810 substance using patients receiving mental health treatment at a VA hospital. Thirty-nine percent of participants in the current study had probable PTSD. This finding is consistent with previous research that has found 10–50% of substance using patients seeking treatment have PTSD (Brown et al., 1995; Driessen et al., 2008; Najavits et al., 2003). Similar to research in non-veteran samples (Chermack et al., 2010; Epstein-Ngo et al., 2014), different factors were associated with different types of violence perpetration. Consistent with our hypotheses, veterans in our sample with probable PTSD were more likely to have reported past-year non-partner physical aggression and non-partner injury aggression. These results indicate that veterans with PTSD and substance problems are at heightened risk for multiple types of violence directed at non-partners. Thus, PTSD might be an important marker for violence involvement among veterans in substance use treatment.
Contrary to our hypotheses, probable PTSD was not significantly associated with partner physical or partner injury violence after adjustment for other variables. This finding was somewhat surprising given previous research connecting PTSD to intimate partner violence in both general (Hahn et al., 2015; Okuda et al., 2015) and veteran samples (Marshall et al., 2005; Taft et al., 2009b; Teten et al., 2010). However, prior research on PTSD and partner violence has generally not been conducted among substance using samples. Thus, the association between PTSD and intimate partner violence might vary by substance use status. Another explanation for inconsistent findings may be measurement and sample differences. For example, prior studies conducted in the general population have measured violence only via six items that include both physical aggression and aggression leading to injury. In addition, prior research among veterans has generally focused on recruiting samples from PTSD clinics. Probable depression was associated with partner injury violence, suggesting that depression may be more related to partner violence than PTSD. Given the lack of research investigating the relationships between different types of violence and both depression and PTSD, further research is needed to tease apart the various aspects of interpersonal violence and their potential connections with these disorders.
Clinically, research on psychotherapy interventions provides evidence that posttreatment substance use is related to posttreatment violence perpetration (Murphy and Ting, 2010; O'Farrell et al., 2003; O'Farrell et al., 2004; Walton et al., 2002); however, it is unknown if aggression is related to posttreatment changes in trauma-related or depressive symptoms. Future research should examine the impact of effective PTSD and depression treatments, such as trauma-focused therapy, cognitive behavioral therapy for depression, and treatment for co-occurring disorders, on interpersonal violence to elucidate whether further treatment development is needed to address both mental health symptomatology and help veterans build skills to reduce violence.
In adjusted models, substance use variables were associated with non-partner violence but not partner violence. Specifically, heavy drinking was associated with non-partner injury violence and cocaine was associated with non-partner physical and injury aggression. These results are somewhat consistent with the findings of Epstein-Ngo and colleagues (2014) in which heavy drinking and cocaine use were associated with non-partner but not partner violence in a substance use treatment sample. Similarly, other studies have found that greater alcohol use is linked to non-partner violence; however, some studies have failed to show that alcohol use is related to partner violence among individuals with substance use or in substance use treatment (Chermack et al., 2010; Alexandercikova et al., 2013). Other research has documented associations between alcohol and partner violence (Chermack et al., 2008; Marshall et al., 2005) and between cocaine and partner violence (Chermack et al., 2010). One explanation may be that within a sample of people seeking treatment for substance use, instead of a more heterogeneous sample, partner violence is more strongly associated with factors other than a specific substance. It is also possible that a more comprehensive measure of problematic alcohol and cocaine use would have yielded different results, given that this study was limited by single items assessing heavy drinking and cocaine use over the prior 30 days. Marijuana use was not associated with any category of violence perpetration in the adjusted models. Although some studies support an association between marijuana use and higher rates of partner violence (Epstein-Ngo et al., 2014), other studies have found no association between marijuana and violence (Chermack et al., 2010; Epstein-Ngo et al., 2012).
Overall, the associations between demographic characteristics (e.g., younger age) and violence perpetration were consistent with prior research (Elbogen et al., 2014; Murray et al., 2008; Stappenbeck et al., 2014). Being married or living with a partner increased the odds of partner physical and injury violence. Consistent with prior research stating that men are more likely to perpetrate non-partner violence (Chermack et al., 2001; Walton et al., 2007) and violence leading to injury (Chermack et al., 2010; Chermack et al., 2014; Murray et al., 2008), the current study found male gender was significantly related to non-partner injury aggression. Probable depression had an inverse relationship with non-partner injury aggression. This finding may reflect the isolative behavior that accompanies symptoms of depression.
Several limitations should be noted. First, the cross-sectional design utilizing measures that assessed symptoms over different time periods prevented us from determining temporal associations between variables and whether variables of interest predated or followed incidences of aggression. Furthermore, no conclusions regarding causation can be made. Second, probable PTSD was determined by self-report perhaps leading to inaccurate classification. Given that self-reports are less accurate, future research should utilize a clinician-administered measure of PTSD to improve diagnostic accuracy. Third, relatively brief measures of substance use were utilized in the current study. More comprehensive measures tend to show stronger relationships between violence and substance use (e.g. Chermack et al., 2010). Fourth, our data included a small sample of women. Future research should include more female veterans. Finally, this study utilized a sample of veterans seeking treatment through the VA which may reduce the generalizability of the findings to other populations.
Despite these limitations, this was the first study to investigate the association between probable PTSD and multiple forms of violence directed at partners and non-partners in a veteran, treatment-seeking sample. The relatively large sample and strong measures of aggression were strengths of the current study. These findings add to the literature by identifying important differences across partner and non-partner violence. Specifically, when accounting for other variables, probable PTSD is associated with only non-partner violence, whereas depression is associated with partner aggression. These findings suggest that PTSD and depressive symptoms are risk factors for violence perpetration among veterans in substance use treatment, and highlight the importance of screening for PTSD, depression, substance use, and both partner and non-partner violence in veterans presenting to treatment. The results also underscore the need for future research to utilize longitudinal designs to disentangle associations between PTSD, depression, substance use, and interpersonal violence. Finally, it will be important to examine which treatments are effective at decreasing not only partner, but also non-partner violence in veterans, and develop and test integrated interventions aimed at reducing both symptom burden and violence.
Highlights.
The study examined PTSD and partner/non-partner aggression in substance using veterans.
Probable PTSD was associated with non-partner aggression at different severity levels.
Probable depression was significantly associated with more severe partner aggression.
Future research is needed to determine what treatments may be effective in decreasing aggression.
Acknowledgments
Role of Funding Source
This study was supported by a Department of Veterans Affairs Office of Research and Development grant IIR 099-333 (Clinicaltrials.gov #NCT01337973) awarded to Dr. Chermack and NIAAA K23AA022641 awarded to Dr. Epstein-Ngo. Dr. Bohnert is supported by a Department of Veterans Affairs HSR&D Career Development Award (CDA 11-245). Additionally, writing of this manuscript was supported by the Department of Veterans Affairs Office of Academic Affiliations, Advanced Fellowship Program in Mental Illness Research and Treatment, the Veterans Affairs Ann Arbor Health Care System, and the Veterans Affairs Serious Mental Illness Treatment Resource and Evaluation Center (SMITREC). The Department of Veterans Affairs and NIAAA had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Author Disclosures
Conflict of Interest
All authors declare that they have no conflicts of interest.
Contributors
Drs. Buchholz, Bohnert, Sripada, and Chermack developed the research question, and designed the analyses. Dr. Buchholz conducted the data analyses and wrote the initial draft of the manuscript. Dr. Chermack designed and directed the parent study. Drs. Rauch and Epstein-Ngo also assisted with designing and conducting the parent study. All authors contributed to drafting and critical revisions of the manuscript. Additionally, all authors have read the manuscript, and they approve of its submission.
References
- Alexandercikova Z, Walton MA, Chermack ST, Cunningham RM, Barry KL, Blow FC. Correlates of partner and non-partner aggression among patients with substance use disorders in an urban ED. Journal of Substance Use. 2013;18:262–274. [Google Scholar]
- Brown PJ, Recupero PR, Stout RL. PTSD substance abuse comorbidity and treatment utilization. Addict. Behav. 1995;20:251–254. doi: 10.1016/0306-4603(94)00060-3. [DOI] [PubMed] [Google Scholar]
- Chermack ST, Grogan-Kaylor A, Perron BE, Murray RL, De Chavez P, Walton MA. Violence among men and women in substance use disorder treatment: a multi-level event-based analysis. Drug Alcohol Depend. 2010;112:194–200. doi: 10.1016/j.drugalcdep.2010.06.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Chermack ST, Murray R, Kraus S, Walton MA, Cunningham RM, Barry KL, Booth BM, Blow FC. Characteristics and treatment interests among individuals with substance use disorders and a history of past six-month violence: findings from an emergency department study. Addict. Behav. 2014;39:265–272. doi: 10.1016/j.addbeh.2013.10.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Chermack ST, Murray RL, Walton MA, Booth BA, Wryobeck J, Blow FC. Partner aggression among men and women in substance use disorder treatment: correlates of psychological and physical aggression and injury. Drug Alcohol Depend. 2008;98:35–44. doi: 10.1016/j.drugalcdep.2008.04.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Chermack ST, Roll J, Reilly M, Davis L, Kilaru U, Grabowski J. Comparison of patient self-reports and urinalysis results obtained under naturalistic methadone treatment conditions. Drug and Alcohol Depend. 2000;59:43–49. doi: 10.1016/s0376-8716(99)00106-4. [DOI] [PubMed] [Google Scholar]
- Chermack ST, Walton MA, Fuller BE, Blow FC. Correlates of expressed and received violence across relationship types among men and women substance abusers. Psychology of Addict. Behav. 2001;15:140–151. doi: 10.1037//0893-164x.15.2.140. [DOI] [PubMed] [Google Scholar]
- Crane CA, Hawes SW, Devine S, Easton CJ. Axis I psychopathology and the perpetration of intimate partner violence. J. Clin. Psychol. 2014;70:238–247. doi: 10.1002/jclp.22013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Driessen M, Schulte S, Luedecke C, Schaefer I, Sutmann F, Ohlmeier M, Kemper U, Koesters G, Chodzinski C, Schneider U, Broese T, Dette C, Havemann-Reinicke U, Group TR-S. Trauma and PTSD in patients with alcohol, drug, or dual dependence: a multi-center study. Alcohol Clin. Exp. Res. 2008;32:481–488. doi: 10.1111/j.1530-0277.2007.00591.x. [DOI] [PubMed] [Google Scholar]
- Elbogen EB, Johnson SC, Wagner HR, Sullivan C, Taft CT, Beckham JC. Violent behaviour and post-traumatic stress disorder in US Iraq and Afghanistan veterans. Br. J. Psychiatry. 2014;204:368–375. doi: 10.1192/bjp.bp.113.134627. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Epstein-Ngo Q, Walton M, Chermack S. Distal and proximal risk factors for non-partner violence in a substance use disorder treatment sample. Alcohol Clin. Exp. Res. 2012;36:231A–231A. [Google Scholar]
- Epstein-Ngo QM, Walton MA, Sanborn M, Kraus S, Blow F, Cunningham R, Chermack ST. Distal and proximal factors associated with aggression towards partners and non-partners among patients in substance abuse treatment. J. Subst. Abuse Treat. 2014;47:282–292. doi: 10.1016/j.jsat.2014.05.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hahn JW, Aldarondo E, Silverman JG, McCormick MC, Koenen KC. Examining the Association between Posttraumatic Stress Disorder and Intimate Partner Violence Perpetration. J. Fam. Violence. 2015;30:743–752. [Google Scholar]
- Harford TC, Yi HY, Grant BF. Other- and self-directed forms of violence and their relationships to DSM-IV substance use and other psychiatric disorders in a national survey of adults. Compr. Psychiatry. 2013;54:731–739. doi: 10.1016/j.comppsych.2013.02.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Heinz AJ, Makin-Byrd K, Blonigen DM, Reilly P, Timko C. Aggressive behavior among military veterans in substance use disorder treatment: the roles of posttraumatic stress and impulsivity. J. Subst. Abuse Treat. 2015;50:59–66. doi: 10.1016/j.jsat.2014.10.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kroenke K, Spitzer RL, Williams JBW. The PHQ-9. J. Gen. Intern. Med. 2001;16:606–613. doi: 10.1046/j.1525-1497.2001.016009606.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Macdonald S, Anglin-Bodrug K, Mann RE, Erickson P, Hathaway A, Chipman M, Rylett M. Injury risk associated with cannabis and cocaine use. Drug and Alcohol Depend. 2003;72:99–115. doi: 10.1016/s0376-8716(03)00202-3. [DOI] [PubMed] [Google Scholar]
- MacManus D, Rona R, Dickson H, Somaini G, Fear N, Wessely S. Aggressive and violent behavior among military personnel deployed to Iraq and Afghanistan: prevalence and link with deployment and combat exposure. Epidemiol. Rev. 2015;37:196–212. doi: 10.1093/epirev/mxu006. [DOI] [PubMed] [Google Scholar]
- Marshall AD, Panuzio J, Taft CT. Intimate partner violence among military veterans and active duty servicemen. Clin. Psychol. Rev. 2005;25:862–876. doi: 10.1016/j.cpr.2005.05.009. [DOI] [PubMed] [Google Scholar]
- Moore TM, Stuart GL, Meehan JC, Rhatigan DL, Hellmuth JC, Keen SM. Drug abuse and aggression between intimate partners: a meta-analytic review. Clin. Psychol. Rev. 2008;28:247–274. doi: 10.1016/j.cpr.2007.05.003. [DOI] [PubMed] [Google Scholar]
- Murphy CM, Ting L. The effects of treatment for substance use problems on intimate partner violence: A review of empirical data. Aggress. Violent Behav. 2010;15:325–333. [Google Scholar]
- Murray RL, Chermack ST, Walton MA, Winters JJ, Booth BM, Blow FC. Psychological aggression, physical aggression, and injury in nonpartner relationships among men and women in treatment for substance-use disorders. J. Stud. Alcohol Drugs. 2008;69:896–905. doi: 10.15288/jsad.2008.69.896. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Najavits LM. Theoretical perspective on posttraumatic stress disorder and substance use disorder. Australian Psychologist. 2005;40:118–126. [Google Scholar]
- Najavits LM, Runkel R, Neuner C, Frank AF, Thase ME, Crits-Christoph P, Blaine J. Rates and symptoms of PTSD among cocaine-dependent patients. J. Stud. Alcohol Drugs. 2003;64:601–606. doi: 10.15288/jsa.2003.64.601. [DOI] [PubMed] [Google Scholar]
- Najavits LM, Weiss RD, Shaw SR. The link between substance abuse and posttraumatic stress disorder in women. Am. J. Addict. 1997;6:273–283. [PubMed] [Google Scholar]
- O'Farrell TJ, Fals-Stewart W, Murphy M, Murphy CM. Partner violence before and after individually based alcoholism treatment for male alcoholic patients. J. Consult. Clin. Psychol. 2003;71:92. doi: 10.1037//0022-006x.71.1.92. [DOI] [PubMed] [Google Scholar]
- O'Farrell TJ, Murphy CM, Stephan SH, Fals-Stewart W, Murphy M. Partner violence before and after couples-based alcoholism treatment for male alcoholic patients: the role of treatment involvement and abstinence. J. Consult. Clin. Psychol. 2004;72:202. doi: 10.1037/0022-006X.72.2.202. [DOI] [PubMed] [Google Scholar]
- Okuda M, Olfson M, Wang S, Rubio JM, Xu Y, Blanco C. Correlates of intimate partner violence perpetration: results from a National Epidemiologic Survey. J. Trauma. Stress. 2015;28:49–56. doi: 10.1002/jts.21986. [DOI] [PubMed] [Google Scholar]
- Owens GP, Held P, Blackburn L, Auerbach JS, Clark AA, Herrera CJ, Cook J, Stuart GL. Differences in relationship conflict, attachment, and depression in treatment-seeking veterans with hazardous substance use, PTSD, or PTSD and hazardous substance use. J. Interpers. Violence. 2014;29:1318–1337. doi: 10.1177/0886260513506274. [DOI] [PubMed] [Google Scholar]
- Pulay AJ, Dawson DA, Hasin DS, Goldstein RB, Ruan WJ, P PR, Huang B, Chou SP, Grant BF. Violent behavior and DSM-IV psychiatric disorders: Results from the National Epidemiologic Survey on Alcohol and Related Conditions. J. Clin. Psychiatry. 2008;69:12–22. doi: 10.4088/jcp.v69n0103. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ruggiero KJ, Ben KD, Scotti JR, Rabalais AE. Psychometric Properties of the PTSD Checklist—Civilian Version. J. Trauma. Stress. 2003;16:495–502. doi: 10.1023/A:1025714729117. [DOI] [PubMed] [Google Scholar]
- Seal KH, Cohen G, Waldrop A, Cohen BE, Maguen S, Ren L. Substance use disorders in Iraq and Afghanistan veterans in VA healthcare, 2001–2010: Implications for screening, diagnosis and treatment. Drug Alcohol Depend. 2011;116:93–101. doi: 10.1016/j.drugalcdep.2010.11.027. [DOI] [PubMed] [Google Scholar]
- Simpson LE, Christensen A. Spousal agreement regarding relationship aggression on the Conflict Tactics Scale-2. Psychol. Assess. 2005;17:423–432. doi: 10.1037/1040-3590.17.4.423. [DOI] [PubMed] [Google Scholar]
- Singh V, Tolman R, Walton M, Chermack S, Cunningham R. Characteristics of men who perpetrate intimate partner violence. J. Am. Board Fam. Med. 2014;27:661–668. doi: 10.3122/jabfm.2014.05.130247. [DOI] [PubMed] [Google Scholar]
- Smith G, Babor T, Burnam M, Mosley C, Rost K, Burns B. Substance abuse outcomes module user’s manual. Little Rock, AR: University of Arkansas for Medical Sciences; 1996. p. b40. [Google Scholar]
- Sontag D, Alvarez L. Across America, deadly echoes of foreign battles. New York: The New York Times; 2008. [Google Scholar]
- Stappenbeck CA, Hellmuth JC, Simpson T, Jakupcak M. The Effects of Alcohol Problems, PTSD, and Combat Exposure on Nonphysical and Physical Aggression Among Iraq and Afghanistan War Veterans. Psychol. Trauma. 2014;6:65–72. doi: 10.1037/a0031468. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Straus MA, Hamby SL, Boney-McCoy S, Sugarman DB. The revised conflict tactics scales (CTS2) development and preliminary psychometric data. J. Fam. Issues. 1996;17:283–316. [Google Scholar]
- Taft CT, Monson CM, Hebenstreit CL, King DW, King LA. Examining the Correlates of Aggression Among Male and Female Vietnam Veterans. Violence Vict. 2009a;24:639–652. doi: 10.1891/0886-6708.24.5.639. [DOI] [PubMed] [Google Scholar]
- Taft CT, Weatherill RP, Woodward HE, Pinto LA, Watkins LE, Miller MW, Dekel R. Intimate partner and general aggression perpetration among combat veterans presenting to a posttraumatic stress disorder clinic. Am. J. Orthopsychiatry. 2009b;79:461–468. doi: 10.1037/a0016657. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Teten AL, Schumacher JA, Taft CT, Stanley MA, Kent TA, Bailey SD, Dunn NJ, White DL. Intimate partner aggression perpetrated and sustained by male Afghanistan, Iraq, and Vietnam veterans with and without posttraumatic stress disorder. J. Interpers. Violence. 2010;25:1612–1630. doi: 10.1177/0886260509354583. [DOI] [PubMed] [Google Scholar]
- Walton MA, Chermack ST, Blow FC. Correlates of received and expressed violence persistence following substance abuse treatment. Drug and Alcohol Depend. 2002;67:1–12. doi: 10.1016/s0376-8716(02)00016-9. [DOI] [PubMed] [Google Scholar]
- Walton MA, Cunningham RM, Chermack ST, Maio R, Blow FC, Weber J. Correlates of violence history among injured patients in an urban emergency department: gender, substance use, and depression. J. Addict. Dis. 2007;26:61–75. doi: 10.1300/J069v26n03_07. [DOI] [PubMed] [Google Scholar]
- Weathers F, Litz B, Herman D, Huska J, Keane T. The PTSD checklist-civilian version (PCL-C) Boston, MA: National Center for PTSD; 1994. [Google Scholar]