Abstract
Preliminary evidence underscores links between attention-deficit hyperactivity disorder (ADHD) symptoms and intimate partner violence (IPV) perpetration and victimization. However, little is known about whether ADHD symptoms are uniquely associated with IPV perpetration and victimization beyond well-established risk factors of IPV commonly associated with the disorder. In a cross-sectional design, 433 college students rated their ADHD symptoms as well as frequencies of psychological and physical IPV perpetration and victimization. Additional risk factors of IPV included childhood maltreatment, primary psychopathy, alcohol abuse, and illicit drug use. Correlational analyses indicated that students with greater ADHD symptom severity reported higher rates of psychological and physical IPV perpetration, and higher rates of psychological IPV victimization. Regression analyses indicated that ADHD symptoms were not additive risk factors of psychological IPV perpetration and victimization. Students reporting any alcohol abuse or illicit drug use endorsed high rates of psychological IPV perpetration and victimization, regardless of their level of ADHD symptoms. However, students who reported no alcohol abuse or drug use, but did report greater ADHD symptom severity—particularly inattention, indicated higher rates of psychological IPV perpetration and victimization than those reporting no alcohol abuse or drug use and low ADHD symptoms. These findings extend prior research by indicating that alcohol abuse and illicit drug use moderate associations between ADHD symptoms and psychological IPV perpetration and victimization. Investigations are needed to identify mechanisms of the association between ADHD symptoms and IPV perpetration and victimization, particularly those abusing alcohol and drugs, for appropriate prevention and intervention efforts to be developed.
Intimate partner violence (IPV), often defined as the use or threat of psychological or physical by a current or former romantic partner (Centers for Disease Control and Prevention [CDC], 2014), is a significant public health concern. Nearly half of all women (48.4%) and men (48.8%) in a recent national investigation reported having experienced psychological aggression (e.g., verbal abuse, coercive control) by an intimate partner in their lifetime (Black et al., 2011). One in four women and one in seven men from the same national study also reported having been victims of severe physical violence (e.g., hit with a fist or something hard, beaten, slammed against something) by an intimate partner in their lifetime. Although the adverse impact of IPV on male victims is beginning to come to light (Black et al., 2011), the deleterious outcomes experienced by female victims are well-established (CDC, 2014). Recent estimates indicate that women are victims of more than five million IPV incidents each year, resulting in nearly two million injuries, and eight million paid work days lost (National Center for Injury Prevention and Control, 2003) as well as a host of negative psychological sequelae (e.g., depression, substance use; Coker et al., 2002).
IPV is especially prevalent among college-aged adults (Capaldi, Knoble, Shortt, & Kim, 2012). According to national estimates, the developmental period when most women (47.1%) and men (38.6%) experience their first victimization occurred is ages of 18 and 24 (Black et al., 2011). During this period, often referred to as “emerging adulthood” (Arnett, 2004), interpersonal relationships assume new meaning as social, emotional, and occupational or educational goals mature. In particular, the ability to successfully initiate romantic relationships (e.g., making a good first impression) is tested as are skills important to sustaining intimate relationships (e.g., compromising, managing conflict). Unfortunately, some emerging adults may lack such skills and resort to violence in interactions with romantic partners.
There are numerous risk factors for emerging adults perpetrating and being victims of IPV (Capaldi et al., 2012). In a recent review, Flynn and Graham (2010) categorized IPV risk variables into three groups: background and personal attributes of perpetrator or victim (e.g., longstanding mental health problems, past experiences with abuse), current lifestyle behaviors (e.g., alcohol and drug use tendencies), and immediate precursors (e.g., provocative acts by partner). One population with chronic mental health difficulties (Barkley, Murphy, & Fischer, 2008), risky lifestyle behaviors (e.g., alcohol use disorders; Lee, Humphreys, Flory, Liu, & Glass, 2011), and sensitivity to partner provocation (Wymbs & Dawson, 2014) are emerging adults with elevated symptoms of attention-deficit hyperactivity disorder (ADHD). Not surprisingly, early evidence highlights that greater ADHD symptom severity increase risk of emerging adults being verbally and physically aggressive with romantic partners (Fang, Massetti, Ouyang, Grosse, & Mercy, 2011; Theriault & Holmberg, 2001; Wymbs et al., 2012). However, important questions regarding the association between ADHD and IPV have yet to be addressed. To this end, the present study investigated whether ADHD symptoms are uniquely associated with IPV perpetration and victimization beyond well-established risk factors for IPV.
ADHD Symptoms as Risk Factors for IPV
Once exclusively considered a disorder of childhood (e.g., American Psychiatric Association [APA], 1994), ADHD symptoms and impairment are now recognized to persist for many into adulthood (APA, 2013). Weyandt and DuPaul (2013) indicated that 2% to 8% of college students have clinically significant ADHD symptoms and impairment, including difficulties with interpersonal relations (e.g., Barkley et al., 2008). Romantic partners of adults with ADHD have identified specific behaviors consistent with symptoms of ADHD, including inattention (e.g., not remembering being told things), hyperactivity/impulsivity (e.g., saying things without thinking), which spark discord in their relationships (Robin & Payson, 2002). In turn, men and women with ADHD, particularly those with elevated inattention and hyperactivity/impulsivity symptoms, report lower romantic relationship satisfaction and have greater difficulties resolving conflicts than those without ADHD (Canu, Tabor, Michael, Bazzini, & Elmore, 2014; Eakin et al., 2004).
Preliminary evidence indicates that ADHD symptoms are risk factors for IPV perpetration and victimization. Theriault and Holmberg (2001) found that male and female college students (Mage = 19.8) reporting greater current ADHD symptoms tended to report elevated rates of physical assault perpetration. Similarly, men and women (Mage = 22) retrospectively reporting greater childhood ADHD symptoms have been found to endorse perpetrating physical IPV as adults more often than those without childhood ADHD symptoms, and the magnitude of this effect did not vary between men and women (Fang et al., 2011). Finally, emerging adult men diagnosed with ADHD in childhood (Mage = 20) reported greater psychological and physical IPV perpetration than those without ADHD histories (Mage = 19.8; Wymbs et al., 2012).
Although compelling, methodological limitations of these studies suggest caution should be taken when interpreting their results. First, in their study of IPV perpetration, Wymbs and colleagues’ (2012) sample did not include women, which is noteworthy given that women perpetrate physical IPV at rates equal to, or slightly higher than, those of men (Capaldi et al., 2012). Second, none of the studies examined whether greater ADHD symptom severity increased risk of psychological or physical IPV victimization. The negative sequelae of victimization (Black et al., 2011; Coker et al., 2002; National Center for Injury Prevention and Control, 2003) highlight the need for researchers to assess whether ADHD symptoms increase risk of being victims of psychological or physical IPV. Addressing these limitations may have important clinical implications, including whether there is a need to develop or adapt gender-specific IPV prevention programs for students with ADHD.
ADHD as a Unique or Moderated Risk Factor
Although preliminary evidence suggests that ADHD symptoms are positively associated with IPV, the degree to which they are uniquely associated with IPV beyond other risk factors of IPV commonly correlated with ADHD remains unclear. For example, childhood conduct problems are often comorbid with ADHD (Barkley, 2006) and robust predictors of IPV perpetration (e.g., Capaldi, Dishion, Stoolmiller, & Yoerger, 2001). Two of the aforementioned studies (Fang et al., 2011; Theriault & Holmberg, 2001) found that ADHD symptoms, particularly the hyperactivity/impulsivity dimension, were additive predictors of IPV perpetration beyond the contribution of childhood conduct problem histories. However, Wymbs et al. (2012) reported that adult ADHD symptoms were not additive predictors of psychological or physical IPV perpetration beyond Antisocial Personality Disorder (APD) symptoms, which are often elevated in adults with ADHD (Barkley et al., 2008) and are robust predictors of IPV perpetration (Holtzworth-Munroe & Stuart, 1994). Taken together, one may speculate that ADHD symptoms are uniquely associated with IPV perpetration beyond past, but not present, conduct problems. Additional tests are needed to extend these findings. Accordingly, recent work has highlighted positive associations between primary psychopathic traits (e.g., callousness, manipulativeness, lack of remorse) and IPV perpetration (e.g., Harris, Hilton, & Rice, 2011). ADHD symptoms have been shown to be positively associated with psychopathy in violent populations (e.g., Langevin & Curnoe, 2010). Nonetheless, studies have yet to investigate whether ADHD symptoms increase risk of perpetration beyond primary psychopathy.
Having experienced verbal, physical, or sexual abuse from parents as children has been shown to predict the rate (Cui, Durtschi, Donnellan, Lorenz, & Conger, 2010; Whitfield, Anda, Dube, & Felitti, 2003), the recurrence (White & Smith, 2004), and the severity of IPV perpetration (Ehrensaft et al., 2003). Histories of childhood maltreatment also increase risk of being a victim of IPV (Renner & Slack, 2006; Whitfield et al., 2003). Recent work has demonstrated that ADHD is positively associated with histories of childhood maltreatment (Briscoe-Smith & Hinshaw, 2006; De Sanctis, Nomura, Newcorn, & Halperin, 2012). To clarify that ADHD is uniquely linked with IPV perpetration and victimization, and underscore that these links are not better explained by associations with childhood maltreatment, studies should control for maltreatment histories when testing ADHD as a risk factor of IPV.
Alcohol abuse and illicit drug use are lifestyle behaviors commonly associated with IPV. Both have been shown to increase risk of perpetrating IPV (Abbey, Wegner, Woerner, Pegram, & Pierce, 2014; Feingold, Kerr, & Capaldi, 2008; Stith, Smith, Penn, Ward, & Tritt, 2004; Thompson & Kingree, 2006) and of being a victim of IPV (Combs-Lane & Smith, 2002; Parks & Fals-Stewart, 2004; Testa, Livingston, & Leonard, 2003). Alcohol abuse and illicit drug use are common among emerging adults with elevated ADHD symptoms (e.g., Lee et al., 2011). It remains necessary to evaluate whether greater ADHD symptom severity is associated with IPV perpetration over and above the risk conferred by substance use and abuse.
Lastly, it is surprising that no prior studies explored whether any of these robust predictors of IPV moderated associations between ADHD symptoms and IPV perpetration or victimization. Adults at greatest risk to perpetrate IPV, or to be victims of IPV, may be those who have greater ADHD symptom severity and well-established background or current lifestyle risk factors (e.g., primary psychopathy, alcohol abuse, illicit drug use, child maltreatment). Conversely, adults reporting greater ADHD symptom severity, but no other risk factors, may be at comparatively less risk of perpetrating or being victims of IPV if ADHD only increases risk of perpetration or victimization in the context of other factors. Studies evaluating whether ADHD symptoms are uniquely associated with IPV perpetration and victimization beyond other robust risk factors of IPV, and those examining whether other robust risk factors moderate links between ADHD symptoms and IPV perpetration and victimization, could lead to enhanced screening procedures for students at risk for IPV on campus.
Study Aims
The first goal of this study was to examine whether ADHD symptoms were associated with IPV perpetration and victimization. We hypothesized that college students with greater ADHD symptom severity would also report perpetrating, and being victims of, greater rates of IPV than those with low ADHD symptoms. In light of prior work indicating that gender does not moderate association between ADHD and IPV (Fang et al., 2011), we also hypothesized that the strength of associations between ADHD symptoms and IPV perpetration and victimization variables would not vary between men and women. The second goal of this study was to investigate whether ADHD symptoms (including separate analyses for inattention and hyperactivity/impulsivity symptoms) were uniquely associated with IPV perpetration and victimization beyond other well-documented risk factors. We hypothesized that ADHD symptoms would be associated with greater risk of perpetrating IPV beyond psychopathic traits, childhood maltreatment, alcohol abuse, and illicit drug use. We also expected that ADHD symptoms would be associated with greater risk IPV victimization beyond childhood maltreatment, alcohol abuse, and illicit drug use. Furthermore, we hypothesized that background or current lifestyle risk factors would moderate associations between ADHD symptoms and levels of IPV perpetration or victimization, such that students with greater ADHD symptom severity and high levels of another risk factor would report greater rates of IPV perpetration and victimization than those with greater ADHD symptom severity-only or those with lower ADHD symptom severity and another risk factor.
Method
Participants
Five hundred and sixty-five 18- to 25-year-old men and women at a Midwestern public university participated in this study. All participants were recruited from the psychology research subject pool and earned one research credit for completing an online survey. We attempted to oversample for individuals with elevated ADHD symptoms by inviting (via email) individuals who reported such symptoms on a subject pool pre-screening measure to participate. Data from 85 participants were excluded for exceeding cutoffs on at least one of the Conner’s Adult Attention-Deficit/Hyperactivity Rating Scale indices suggesting non-credible or invalid responding (see below). An additional participant was excluded as an outlier for endorsing IPV levels greater than 3 SDs above the norm on all psychological aggression and physical assault perpetration and victimization scales. Finally, 46 participants were excluded because they did not provide any IPV perpetration or victimization data. Relative to individuals excluded from the study, included participants were more likely to be female and to have dated someone in the past year.
Among those eligible and with complete data for this study, our sample was comprised of 433 adults whose median age was 19 years (M = 19.10, SD = 1.28). More than half of the participants were female (62.8%), most were Caucasian (87.2%), and most were also exclusively heterosexual (87.5%). Most participants reported living in dormitories or residence halls (77.6%), having dated at least one person in the past year (82.1%), and having been physically intimate or “hooked up” with someone (e.g., kissing, oral sex, intercourse) in the past year (93.7%). Eleven percent of our sample reported previously being diagnosed with ADHD, and 9.1% reported currently being prescribed medication to treat ADHD.
Measures
Participants completed all of the following questionnaires as part of a single online survey. Prior to completing this survey, participants were required to provide informed consent with a digital signature. This study was approved by the University Social-Behavioral Institutional Review Board.
ADHD symptoms
To assess for the presence of ADHD symptoms, we used the Conner’s Adult Attention-Deficit/Hyperactivity Rating Scale Self-Report Long Form (CAARS; Conners, Erhardt, & Sparrow, 1999). The CAARS is a 66-item self-report inventory of ADHD, including Diagnostic and Statistical Manual of Mental Disorders (4th ed.; DSM-IV; APA, 1994) symptoms of inattention and hyperactivity/impulsivity. Responses are scored on a 4-point scale from 0 (not at all) to 3 (very much). The CAARS has strong test-retest reliability (1 month r = .89; Erhardt, Epstein, Conners, Parker, & Sitarenios, 1999) and internal consistency (α = .86-.92; Conners, Erhardt, Epstein, et al., 1999). CAARS scores are positively correlated with other self-report ADHD measures and show good sensitivity and specificity to adult ADHD diagnoses made via semi-structured clinical interview (Erhardt et al., 1999). As ADHD symptoms are often exaggerated by college students (Harrison, Edwards, & Parker, 2007; Sullivan, May, & Galbally, 2007), we used three indices to identify and exclude potentially invalid over-reporters: t scores > 80 and high scores on an inconsistency index developed by Conners, Erhardt and Sparrow (1999), and high scores on an infrequency index developed by Suhr, Buelow, and Riddle (2011). Students exceeding cutoffs on any of these indices were considered non-credible reporters and were excluded from analyses. For this study, average-item scores for total DSM-IV ADHD symptoms (α = .85), inattention symptoms (α = .83), and hyperactivity/impulsivity symptoms (α = .71) were used as risk variables.
Alcohol abuse
We used the hazardous consumption scale of the Alcohol Use Disorders Identification Test (AUDIT-C; Saunders, Aasland, Babor, De La Fuente, & Grant, 1993) to measure alcohol abuse. The AUDIT-C has well-established psychometric properties, including a high degree of test-retest reliability (over a 3- to 4-week interval r = .98; Reinert & Allen, 2007). Internal consistency of the AUDIT-C in this study was acceptable (α = .80). Total scores of five for men and four for women on the AUDIT-C are recommended cutoffs for individuals likely to endorse alcohol use disorders (Reinert & Allen, 2007). In this study, 56.4% of men and 62.2% of women reported hazardous consumption in excess of these cutoffs.
Illicit drug use
Participants also reported on their use of illicit drugs (i.e., marijuana, inhalants, hallucinogens, cocaine, opiates, tranquilizers, MDMA, methamphetamine, club drugs, and misuse of prescription drugs). Similar to our coding of alcohol abuse, we created a dichotomous illicit drug use variable indicating whether or not participants used any illicit drugs in the past year. In this study, 48.6% of women and 55.5% of men endorsed using at least one illicit drug in the past year.
Childhood maltreatment
The Childhood Trauma Questionnaire (CTQ; Bernstein et al., 2003) was used to screen for childhood maltreatment histories. Respondents indicated whether they experienced behaviors consistent with emotional, physical, or sexual abuse by the age of 12 on a scale from 1 (never true) to 5 (very often true). For this study, we collapsed the emotional, physical, and sexual abuse items into one composite index of childhood maltreatment. This child maltreatment composite had acceptable internal consistency (α = .80).
Psychopathy
To measure psychopathic traits, we administered the Levenson Self-Report Psychopathy Scale (LSRP; Levenson, Kiehl, & Fitzpatrick, 1995). Items are rated on a scale from 1 (strongly disagree) to 4 (strongly agree). Evidence supports the construct and convergent validity (Sellbom, 2011) and the test-retest reliability (over an 8-week interval r = .83; Lynam, Whiteside, & Jones, 1999) of the LSRP. Given the symptom overlap between ADHD and secondary psychopathy, only the primary psychopathy subscale (α = .79) measuring pure psychopathy (i.e., selfishness, callousness, and manipulative behavior) was used in this study. Primary psychopathy was explored as a risk factor of IPV perpetration but not victimization, of which it is not theoretically implicated as a risk factor.
IPV perpetration and victimization
The Conflict Tactics Scale–II (CTS-II; Straus, Hamby, Boney-McCoy, & Sugarman, 1996) is the most widely used measure of IPV perpetration and victimization. The CTS-II is reliable with college students (α = .79 to .86; Straus et al., 1996). We focused on examining risk factors of psychological aggression perpetration (α = .77; e.g., I shouted or yelled at my partner) and victimization (α = .77; e.g., My partner shouted or yelled at me) as well as physical assault perpetration (α = .79; e.g., I pushed or shoved my partner) and victimization (α = .77; e.g., My partner pushed or shoved me). The response scale for each item ranges from 0 (this has never happened) to 6 (this has happened more than 20 times in the past year). To ease interpretation, average item scores were computed for all scales.
Results
ADHD Symptoms and IPV Perpetration and Victimization
Descriptive statistics for risk and criterion variables are presented in Table 1. In addition, 67.9% and 69.4% of the students reported perpetrating, or being victims of, at least one psychological IPV behavior in the past year, respectively. Fewer students reported perpetrating (22.1%) or being victims (19.5%) of at least one physical IPV behavior in the past year. Bivariate correlations were computed to determine the degree to which ADHD symptoms (and covariates) were associated with IPV perpetration and victimization (Table 1). ADHD symptoms were significantly, albeit modestly, associated with psychological and physical perpetration as well as with psychological victimization, such that students reporting greater symptoms tended to also report higher levels of perpetration or victimization. ADHD symptoms were not significantly associated with physical victimization. As such, regression models were not conducted with this outcome variable.
Descriptive Statistics and Intercorrelations for Risk and Criterion Variables in Regression Analyses.
Note. Gender (−1 = female, 1 = male); child abuse = average item score for emotional, physical, and sexual abuse as measured by Childhood Trauma Questionnaire (1 = never true to 5 = very often true; Bernstein et al., 2003); primary psychopathy = average item score for primary psychopathy subscale of the Levenson Self-Report Psychopathy Scale (1 = strongly disagree to 4 = strongly agree; Levenson, Kiehl, & Fitzpatrick, 1995); alcohol abuse (−1 = did not exceed clinical cutoff for hazardous alcohol consumption on AUDIT-C; 1 = exceeded cutoff for hazardous alcohol consumption on AUDIT-C; Illicit Drug Use (−1 = did not endorse using any illicit drugs in past year; 1 = endorsed using at least 1 illicit drug in past year); ADHD symptoms = average item score on the Conner’s Adult Attention-Deficit/Hyperactivity Rating Scale (0 = not at all to 3 = very much; Conners et al., 1999); psychological = average item score on the psychological aggression perpetration and victimization scales of the Conflict Tactics Scale–II (0 = this has never happened to 6 = this has happened more than 20 times in the past year; Straus, Hamby, Boney-McCoy, & Sugarman, 1996); physical = average item score on the physical assault perpetration and victimization scales of the Conflict Tactics Scale–II (0 = this has never happened to 6 = this has happened more than 20 times in the past year; Straus et al., 1996). ADHD = attention-deficit hyperactivity disorder; AUDIT = Alcohol Use Disorders Identification Test.
p < .05. **p < .01.
Testing Moderation by Gender
Regression analyses were conducted to explore whether gender moderated associations between ADHD symptoms and IPV perpetration and victimization variables. Mplus 7.0 and the maximum likelihood estimation robust (MLR) to potential non-normality (Muthén & Muthén, 2013) was used to conduct these analyses. All predictors and outcomes were treated as observed variables. We found that gender did not moderate links between ADHD symptoms and IPV perpetration or victimization variables. Thus, the strength of association between ADHD symptoms and IPV perpetration and victimization was similar for men and women. There was a main effect of gender beyond ADHD symptoms in relation to psychological IPV perpetration (β = −.10, SE = .03, p = .003). Regardless of ADHD symptoms, women (75.7%) reported higher rates of psychological IPV perpetration than men (54.7%). As such, gender was included as a covariate in subsequent analyses with this outcome and, to be consistent, with the other criterion variables.
ADHD Symptoms as Risk Factors for IPV
Hierarchical regression analyses were conducted to explore (1) whether ADHD symptoms were associated with psychological and physical perpetration as well as psychological victimization beyond well-established risk factors for IPV (additive models) and (2) whether other risk factors moderated associations between ADHD symptoms and IPV perpetration and victimization variables (moderated models). Mplus 7.0 and the MLR estimator were once again used to conduct these analyses. As before, all predictors and outcomes were treated as observed variables. Analyses were conducted in two steps. Step 1 tested additive models: ADHD was included as a risk factor of IPV perpetration and victimization alongside gender, childhood maltreatment, alcohol use, and primary psychopathy (perpetration only). Step 2 tested moderated models: Interactions between ADHD symptoms and the background or current lifestyle risk factors were included one at a time to assess for unique associations with IPV perpetration and victimization. When ADHD symptoms were significant additive or moderated risk factors, secondary analyses explored whether inattention or hyperactivity/impulsivity symptoms were additive or moderated risk factors of IPV perpetration or victimization.
Perpetration
Table 2 presents the results of the regression analyses exploring risk factors of psychological and physical IPV perpetration. Results of the additive model (Step 1) with psychological aggression indicated that ADHD symptoms were not uniquely associated with psychological IPV perpetration beyond other risk factors. Gender, childhood maltreatment, and alcohol abuse were each associated psychological perpetration, such that young women, hazardous drinkers, and those reporting elevated levels of childhood maltreatment endorsed perpetrating higher rates of psychological IPV than men or students reporting no alcohol abuse or low levels of childhood maltreatment. Results of the moderated models (Steps 2a-e) revealed that illicit drug use moderated associations between ADHD symptoms and psychological IPV perpetration. Students reporting illicit drug use endorsed high rates of psychological IPV perpetration regardless of their level of ADHD symptoms, and those who reported no drug use but greater ADHD symptom severity endorsed higher rates of psychological perpetration than those reporting no drug use and low ADHD symptoms (Figure 1). Secondary analyses found that this interaction was only significant with inattention symptoms (β = −.10, SE = .05, p = .035). The pattern of results was consistent with those presented in Figure 1, such that students reporting any illicit drug use, regardless of ADHD symptoms, and those reporting greater inattention symptom severity only endorsed perpetrating higher rates of psychological IPV than those reporting no drug use and lower inattention symptoms.
Regression Analyses With Psychological and Physical IPV Perpetration.
Note. The values presented in the table are β (SE). IPV = intimate partner violence; ADHD = attention-deficit hyperactivity disorder.
p < .05. **p < .01.

Illicit drug use moderates the association between ADHD symptoms and psychological aggression perpetration.
Results of the additive models with physical IPV also indicated that ADHD symptoms were not uniquely associated with perpetration beyond the other risk factors. Students reporting elevated levels of child maltreatment or greater psychopathic traits endorsed perpetrating higher rates of physical assault than those reporting low maltreatment or low psychopathic traits. Moreover, results of each moderated model revealed that no other risk factors moderated associations between ADHD symptoms and any physical IPV perpetration.
Victimization
Table 3 presents the results of the regression analyses exploring risk factors of psychological victimization. Results of the additive model (Step 1) with psychological aggression indicated that ADHD symptoms did not confer additional risk of students being victims of psychological IPV beyond other risk factors. Only childhood maltreatment and alcohol abuse were uniquely associated with psychological aggression victimization, such that those reporting elevated levels of maltreatment and any hazardous drinking endorsed being a victim of higher rates of psychological aggression than those reporting low levels of maltreatment or no alcohol abuse. Results of moderated models (Steps 2a-d) indicated that alcohol abuse and illicit drug use moderated associations between ADHD symptoms and psychological IPV victimization. Students reporting hazardous alcohol consumption reported high levels of psychological IPV victimization regardless of self-reported ADHD symptoms, and those reporting no alcohol abuse but greater ADHD symptom severity endorsed higher rates of psychological victimization than those reporting no alcohol abuse and lower ADHD symptoms (Figure 2). Secondary analyses found that the interaction was only significant with inattention symptoms (β = −.11, SE = .05, p = .040). Consistent with the pattern of results presented in Figure 2, students reporting any alcohol abuse reported higher rates of psychological IPV victimization regardless of their level of ADHD symptoms, and those reporting no alcohol abuse but endorsing greater inattention symptom severity indicated higher levels of victimization than those reporting no alcohol abuse and lower inattention symptoms. Somewhat differently, students reporting any illicit drug use, and those reporting no drug use but greater ADHD symptom severity, endorsed higher levels of psychological IPV victimization than those reporting no drug use and lower ADHD symptoms (Figure 3). Secondary analyses found that the interaction was significant with inattention symptoms (β = −.13, SE = .05, p = .011) and hyperactivity/impulsivity symptoms (β = −.12, SE = .04, p = .006). The pattern of results was similar to those presented in Figure 3, such that students reporting any illicit drug use, and those reporting no drug use but greater inattention or hyperactivity/impulsivity symptom severity, endorsed higher rates of psychological IPV victimization than students reporting no drug use and lower inattention or hyperactivity/impulsivity symptoms.
Regression Analyses With Psychological IPV Victimization.
Note. The values presented in the table are β (SE). IPV = intimate partner violence; ADHD = attention-deficit hyperactivity disorder.
p < .05. **p < .01.

Alcohol abuse moderates the association between ADHD symptoms and psychological aggression victimization.

Illicit drug use moderates the association between ADHD symptoms and psychological aggression victimization.
Discussion
The present study investigated the degree to which ADHD symptoms are associated with risk of college students perpetrating or being victims of IPV. Correlational analyses indicated that students reporting greater ADHD symptom severity endorsed higher rates of psychological and physical IPV perpetration, as well as higher rates of psychological victimization. These findings replicated the results of Wymbs and colleagues (2012), who found that individuals with ADHD histories were more likely to perpetrate psychological and physical IPV than those without histories. This is the first study to demonstrate that individuals with greater ADHD symptom severity are at risk for being victims of psychological IPV. We also extended the evidence base by highlighting ADHD as a risk factor for IPV in a sample of college student men and women. That said, we did not find that greater ADHD symptom severity was associated with being a victim of physical IPV. Given that ADHD was associated with psychological victimization, finding that it was not associated with physical victimization may have been a spurious result.
Consistent with prior work (Fang et al., 2011), gender did not moderate the association between ADHD symptoms and any IPV perpetration. That is, the strength of associations between ADHD symptoms and psychological and physical IPV perpetration did not differ between men and women in this sample. The correlation between ADHD symptoms and psychological victimization was also similar in strength across men and women. Thus, greater ADHD symptom severity appears to increase rates of IPV perpetration and victimization regardless of the gender of the perpetrator or victim. In addition to replicating these findings, future studies should include men and women in their samples to evaluate whether mechanisms underlying associations between ADHD symptoms and IPV perpetration and victimization are gender-specific, as prior studies have illustrated in college samples with lower ADHD symptoms (e.g., Kelley, Edwards, Dardis, & Gidycz, 2015).
Although men and women reporting greater ADHD symptom severity were at risk of perpetrating and being victims of IPV, ADHD symptoms were not uniquely associated with any perpetration or victimization variable beyond the risk associated with well-established predictors of perpetration and victimization (i.e., childhood maltreatment, alcohol abuse, illicit drug use, and psychopathy). That said, study findings indicated that ADHD symptoms operated as interactive risk factors of IPV perpetration and victimization in three instances. Illicit drug use was found to moderate the association between ADHD symptoms and psychological IPV perpetration and victimization, such that students reporting greater ADHD symptom severity—especially inattention—and no drug use reported nearly the same levels of psychological perpetration, and the same levels of psychological victimization, as those reporting any illicit drug use. This pattern of findings is similar to the results of a study by Weafer, Camarillo, Fillmore, Milich, and Marczinski (2008), who found that sober drivers with elevated ADHD symptoms were just as impaired behind the wheel as intoxicated drivers without ADHD. Investigations are warranted to examine whether students with greater ADHD symptom severity who are sober during interactions with their romantic partner are just as likely to perpetrate, or be victims of, psychological IPV as students with lower ADHD symptoms using illicit drugs during interactions with their partners.
Alcohol abuse was found to moderate the association between ADHD symptoms and psychological aggression victimization. Specifically, students reporting hazardous alcohol consumption reported high levels of psychological victimization, regardless of their level of ADHD symptoms, whereas those reporting no alcohol abuse but greater ADHD symptom severity endorsed more psychological victimization than those reporting no abuse and low ADHD symptoms. This pattern of results not only underscores the relevance of alcohol abuse as a risk factor for IPV victimization, as has been shown previously (e.g., Parks & Fals-Stewart, 2004), but also that ADHD symptoms play a unique role in exacerbating risk of victimization, at least among students not reporting hazardous alcohol consumption. Notably, greater symptoms of inattention, not hyperactivity/impulsivity, increased vulnerability to psychological aggression among those not drinking excessively. As evidence indicates that perpetrators are looking for vulnerable victims (e.g., Lisak & Miller, 2002), perhaps exhibiting symptoms of inattention in social situations (e.g., missing or misinterpreting social cues) makes individuals (appear) more susceptible to psychological aggression.
Although the present study had several strengths (e.g., inclusion of men and women, exclusion of individuals who appeared to be invalid reporters of ADHD symptoms), there were also a number of limitations. First, our results may not generalize to emerging adults not seeking university degrees. While many college students have ADHD (Weyandt & DuPaul, 2013), most adults with elevated ADHD symptoms do not attend college (Kuriyan et al., 2013). Future studies should consider sampling emerging adults who are not exclusively attending college. Second, the amount of variance accounted for in our regression analyses was limited. Researchers interested in replicating these analyses should consider including immediate precursors of IPV (e.g., relationship satisfaction, communication behaviors) in risk models, and oversampling for adults with ADHD diagnosed in childhood, as both strategies may allow for a greater degree of variability to be accounted for. Third, all risk and criterion variables were measured via self-report, which means that reporter bias may have explained a portion of all findings discussed herein. This is concerning given the propensity for individuals to under-report IPV perpetration (e.g., Arias & Beach, 1987) and for emerging adults with greater ADHD symptom severity to inflate self-appraisals (e.g., Knouse, Bagwell, Barkley, & Murphy, 2005). Future studies should solicit information from reliable external informants to confirm ADHD symptoms in childhood (e.g., parents) and in adulthood (e.g., romantic partners). Fourth, all risk and criterion variables were measured at the same time point. Investigations collecting data prospectively would allow for greater confidence in the unique contribution of ADHD symptoms toward prediction of IPV perpetration and (re)victimization outcomes. Fifth, because we did not measure the rate of IPV perpetration or victimization when under the influence of alcohol or drug use, we cannot conclude that using substances contributed directly to IPV rates. Future studies should assess whether individuals with greater ADHD symptom severity are especially vulnerable to IPV perpetration and victimization when using alcohol or illicit drugs. Sixth, because we did not hypothesize that any specific drugs or any specific form of childhood maltreatment would moderate associations between ADHD symptoms and IPV perpetration or victimization variables, illicit drug use, and childhood maltreatment were composite variables formed by collapsing across use of any drugs and across ratings of emotional, physical, or sexual abuse in childhood, respectively. Researchers interested in extending our research may wish to assess for whether use of specific drugs or forms of maltreatment strengthens or weakens associations between ADHD symptoms and IPV perpetration and victimization variables. Finally, we did not examine whether associations between risk factors and IPV varied depending on the severity of IPV reported. It is recommended that researchers attempt to replicate this study extend it by examining whether the strength of associations between ADHD symptoms and levels of IPV perpetration and victimization varies for minor or severe IPV behaviors.
Taken together, findings from this study indicate that college students with greater ADHD symptom severity, particularly those also reporting hazardous alcohol use and use of illicit drugs, are at increased risk of perpetrating and being victims of IPV. In the event findings are replicated, men and women reporting elevated ADHD symptoms may benefit from primary and secondary IPV prevention programs (see Cornelius, 2007 for a review). However, because it is unclear whether any mechanisms unique to ADHD increase risk of IPV perpetration or victimization, it remains to be seen whether standard prevention programs need to be tailored to meet the needs of individuals with greater ADHD symptoms. For example, students with ADHD could benefit more from programs refined to focus on how inattentiveness disrupts social information processing of IPV situational variables (e.g., early identification of risky situations, attention to environmental risk cues). That said, because our findings also suggested that students with greater ADHD symptom severity may be at risk of IPV perpetration and victimization owing to other risk factors (i.e., hazardous drinking and illicit drug use), those designing prevention programs should consider highlighting the dangers of alcohol abuse and illicit drug use for all students, but especially those with greater ADHD symptoms.
Footnotes
Authors’ Note
Parts of this article were presented at the 2013 annual conference of the Association for Behavioral and Cognitive Therapies in Nashville, TN, and at the 2014 annual convention of the American Psychological Association in Washington, DC.
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.
