|Home | About | Journals | Submit | Contact Us | Français|
A substantial body of research supports a strong cross-sectional and longitudinal association between substance misuse and perpetration of intimate partner violence (IPV). This article briefly addresses the theoretical connection between substance use and intimate partner violence and research on the association between substance misuse and IPV. Studies examining the effect of individual and couples-based addiction treatments on IPV are reviewed. The implications of this work and future directions for research are discussed.
Intimate partner violence (IPV) is a major public health concern. Schafer, Caetano, and Clark (1998) surveyed United States couples and found that over 20% reported experiencing one or more episodes of IPV within the past year. According to the National Center for Injury Prevention and Control (2003), at least 1.3 million women are severely victimized each year in the US. Both genders commit high rates of IPV. A meta-analysis of IPV studies demonstrated that women are slightly more likely to engage in at least one act of physical aggression than men; however, male-to-female violence has more detrimental effects than female-to-male violence (see Archer, 2000 for review). For example, victimized women are more likely than victimized men to experience physical injuries and depressive symptomatology and to require medical attention for their injuries (Cascardi, Langhinrichsen, & Vivian, 1992; Tjaden & Thoennes, 2000). Victimized women are also more likely to miss work and use mental health and justice system services than their victimized male counterparts (Stets & Straus, 1990; Tjaden & Thoennes, 2000).
An extensive theoretical and empirical literature links substance misuse to IPV perpetration and victimization (Leonard, 1993; 2001; 2005; Quigley & Leonard, 2000a; Stuart, O’Farrell, & Temple, in press). The relationship between alcohol and IPV has been established cross sectionally and longitudinally (e.g., Chermack & Giancola, 1997). Substance misuse is over-represented in samples seeking treatment for IPV (e.g., Brannen & Rubin, 1996; Brown, Werk, Caplan, & Seraganian, 1999; Dalton, 2001; Stuart, Moore, Kahler, & Ramsey, 2003a; Stuart, Moore, Ramsey, & Kahler, 2003b; 2004; Stuart et al., 2006), and IPV is over-represented in samples seeking treatment for substance misuse (Brown, Werk, Caplan, Sheilds, & Seraganian, 1998; O’Farrell & Murphy, 1995; O’Farrell, Van Hutton, & Murphy, 1999; Stith, Crossman, & Bischof, 1991). Moreover, research has demonstrated a temporal connection between substance use and episodes of IPV (Fals-Stewart, 2003).
This article briefly reviews some theoretical explanations for the connection between substance misuse and IPV, as well as empirical studies highlighting this link. Subsequently, we examine the impact of substance related treatments on IPV, and make suggestions for future work in this area.
Chermack and Giancola (1997) conducted an extensive review of the relationship between alcohol use and aggression. They presented an integrated biopsychosocial theoretical model in which biochemistry, pharmacology and metabolism, gender, provocation and threats, cognition, expectancies, executive functioning, personality, psychopathology, and level of intoxication interact to impact the connection between alcohol and violence. Similarly, Leonard and colleagues (Leonard, 1993; 2005; Leonard & Jacob, 1988; Leonard & Roberts, 1996; Leonard & Senchak, 1996) proposed a theoretical model that represents an integration of the research of past theorists (e.g., Conner & Ackerly, 1994; Murphy & O’Farrell, 1996; O’Leary, 1988; Parsons & Nixon, 1996; Pernanen, 1976; Steele & Josephs, 1990), which emphasized that alcohol may affect IPV via interactions among psychological/cognitive, physiological, and contextual/situational factors. Leonard et al. proposed that alcohol use by both relationship partners may contribute to IPV, hypothesizing that patterns of alcohol use influence IPV over the long term, while acute intoxication may influence particular violent events. Theorists (e.g., Leonard, 2005; Crowell & Burgess, 1996) have noted that the etiology of IPV is likely multifactorial, with a dynamic interplay of variables that influence the likelihood of IPV differently across time and situations.
Leonard theorized that distal influences (e.g., temporally stable individual and couple characteristics, personality traits, male drinking patterns, female drinking patterns, relationship discord), combined with proximal and contextual factors, such as situational cues and acute alcohol influences on both partners, may elicit IPV when in the context of escalating partner conflict. The distal influences in Leonard’s model are thought to be relatively stable over time for both partners in the relationship, meaning that they may be consistently present and provide a background or context in which relationship conflict may arise. Such distal influences might include antisocial personality traits, trait anger, hostility, hazardous drinking patterns of each partner, gender roles, perceived power inequity, and relationship discord (Leonard 1993; 2001; Leonard & Senchak, 1996). The distal factors are proposed to influence more proximal variables, which are episodic and time limited. Proximal variables might include the cognitive and pharmacological impact of acute alcohol consumption, perceptual and appraisal processing of both partners, the social and physical environmental cues in which the conflict occurs, and transient affective states such as state anger. The model proposes that, over time, there may be reciprocal influences between distal and proximal variables; however, Leonard (1993) suggested that it is most useful to view distal variables as strongly influencing proximal variables, which then may lead to IPV.
The theories proposed by Chermack and Giancola (1997), and that of Leonard and colleagues, have been extended to include the influence of drug use on IPV (Moore & Stuart, 2004; 2005; Stuart et al., 2008). However, since different drugs have widely varying physiological and behavioral effects, it is likely that specific substances also have varying effects on IPV perpetration and victimization (Chermack & Blow, 2002; Stuart et al., 2008).
Leonard (2005) described the controversy regarding whether alcohol use should be considered a causal factor in IPV perpetration. He noted that no single study would provide indisputable evidence for a causal role for alcohol use; however, on the basis of converging evidence across multiple domains (some of which is reviewed in this manuscript), he concluded that heavy alcohol consumption is a contributing cause of IPV. Of course, alcohol is neither necessary nor sufficient to elicit IPV (Leonard, 2001).
Empirical studies have revealed that men exhibiting IPV are substantially more likely than a wide variety of comparison groups to use and misuse alcohol and drugs (e.g., Quigley & Leonard, 2000a; Holtzworth-Munroe, Bates, Smutzler, & Sandin, 1997). The link between IPV and alcohol problems and/or alcohol diagnoses has been reported in violent couples in treatment (e.g., Telch & Lindqust, 1984; Rosenbaum & O’Leary, 1981; O’Farrell, Murphy, Stephen, Fals-Stewart, & Murphy, 2004), partner violent men in the community (e.g., Caetano, McGrath, Ramisetty-Mikler, & Field, 2005; Cunradi et al., 1999; Fagan, Barnett, & Patton, 1988; Julian & McKenry, 1993; Kyriacou et al., 1999; Schafer, Caetano, & Cunradi, 2004; Stuart & Holtzworth-Munroe, 2005; Van Hasselt, Morrison, & Bellack, 1985), premarital samples (e.g., Heyman, O’Leary, & Jouriles, 1995; Leonard & Senchak, 1993; 1996; Quigley & Leonard, 1999), military samples (e.g., Pan, Neidig, & O’Leary, 1994), and emergency department samples (e.g., Grisso et al., 1999; Kyriacou et al., 1999; Lipsky, Caetano, & Field, 2005). In addition, a relationship between alcohol use and IPV has been shown in nationally representative samples (e.g., Field & Caetano, 2004; Kantor & Straus, 1987; Leonard & Blane, 1992), community samples (e.g., Fagan et al., 1988; McKenry, Julian, & Gavazzi, 1995; Stuart & Holzworth-Munroe, 2005), premarital samples (e.g., Leonard & Senchak, 1993; Quigley & Leonard, 1999, 2000b), and emergency department populations (e.g., Kyriacou et al., 1999; Lipsky, Caetano, Field, & Larkin, 2004). Further, this association has been demonstrated both cross-sectionally and longitudinally, even after controlling for past IPV (Caetano et al., 2005; Field & Caetano, 2004; Leonard & Senchak, 1996; Quigley & Leonard, 2000a,b; Pan et al., 1994).
Research has shown a strong association between illicit drug use and IPV, particularly among male perpetrators (Chase, O’Farrell, Murphy, Fals-Stewart, & Murphy, 2003; Coker, Smith, McKeown, & King, 2000; Holtzworth-Munroe, Meehan, Herron, Rehman, & Stuart, 2000; Logan, Walker, & Leukefeld, 2001; Testa, Livingston, & Leonard, 2003), even after controlling for antisocial personality (Murphy, O’Farrell, Fals-Stewart, & Feehan, 2001) and alcohol use (Moore & Stuart, 2004; Stuart et al., 2008). Relatively few studies have been conducted on the connection between IPV and specific illicit drugs (El-Bassel, Gilbert, Wu, Go, & Hill, 2005; Stuart et al., 2008; Testa et al., 2003). The majority of this research has examined cocaine and cannabis. Although the experimental literature on the association between cocaine use and aggression toward strangers is equivocal (Hoaken & Stewart, 2003), the majority of studies on cocaine and IPV have found a strong positive relationship (e.g., Bennet, Tolman, Rogalski, & Srinivasaraghavan, 1994; Grisso et al., 1999; Murphy et al., 2001; Parrott, Drobes, Saladin, Coffey, & Dansky, 2003; Stuart et al., 2008). Moreover, research has shown that cocaine use is associated with severe IPV perpetration and victimization in men and women (Chermack, Fuller, & Blow, 2000; Chermack, Walton, Fuller, & Blow, 2001).
The experimental literature on acute marijuana use and aggression is mixed. Although there is evidence that higher amounts may reduce aggression (Cherek et al., 1993; Myerscough & Taylor, 1985), cannabis withdrawal may facilitate violence (Hoaken & Stewart, 2003; Moore & Stuart, 2005). In addition, cross sectional (Buttell & Carney, 2006; Chermack et al., 2000; 2001; Logan et al., 2001; Moffitt, Caspi, Harrington, & Milne, 2002; Murphy et al., 2001) and longitudinal (Testa et al., 2003; Woffordt, Mihalic, & Menard, 1994; but see Fals-Stewart, Golden, and Schumacher, 2003) research has shown a positive relationship between marijuana use and IPV. For example, in a longitudinal study, El-Bassel et al. (2005) found that women in methadone maintenance treatment who used marijuana were over four times more likely to experience IPV than women who did not use marijuana.
An increasing body of research has demonstrated that men and women participating in batterer intervention programs (i.e., individuals arrested for domestic violence and court-referred to violence prevention programs) misuse alcohol and drugs at a high rate. Dalton (2001) reported that 49% of a sample composed of men arrested for IPV “had an indication of substance abuse” and Brannen and Rubin (1996) found that approximately one-quarter of their sample of arrested men endorsed a drinking problem on their screening intake form. Gondolf (1999) reported that 56% of the court-referred male batterers in his study scored in the “alcoholic tendencies” range, and 26% had a history of some form of treatment for substance misuse. Stuart et al. (2003a) found that over two-thirds of a sample of 150 male batterers met criteria for hazardous drinking. Half of the male batterers had a probable alcohol-related diagnosis and approximately one-third reported symptoms consistent with a drug-related diagnosis. Across 2 studies of women arrested for domestic violence (Stuart et al., 2003b; 2004), approximately half of the women drank hazardously, approximately one-third of the women reported symptoms consistent with an alcohol abuse or dependence diagnosis, and approximately one-quarter reported symptoms consistent with a drug-related diagnosis. Further, in a study of 271 men and 135 women arrested for violence, Stuart et al. (2008) found that over half of male and female perpetrators had used illicit drugs in the past year.
Fals-Stewart (2003) highlighted the temporal association between alcohol and drug use in a sample of men in a domestic violence intervention program and in a sample of men entering an alcoholism treatment program. In both samples, IPV was 8-19 times more likely to occur on a drinking day than on a non-drinking day. Another study revealed that male-to-female IPV was three times more likely to occur on a day that the male partner used cocaine compared to a non-use day, after controlling for antisocial personality and relationship discord (Fals-Stewart et al., 2003).
Additional evidence for the link between substance misuse and IPV can be derived from the batterer intervention outcome literature, in which individuals with substance misuse have an increased risk for violence recidivism relative to individuals without substance misuse (e.g., DeMaris & Jackson, 1987; Hamberger & Hastings, 1990; Stuart, 2005; Stuart, Temple & Moore, 2007a). It should be noted that the efficacy of batterer intervention programs is questionable, irrespective of batterers’ substance use. In a meta-analysis of 22 batterer intervention studies in which a comparison group was included (e.g., program drop-outs, nonequivalent controls, etc.), Babcock, Green, and Robie (2004) found a small overall effect (d=.18). Recidivism was 5% less likely by men arrested and referred to a batterer intervention program than by men arrested and sanctioned without intervention. Similarly, Feder and Wilson (2005) conducted a meta-analysis using only the 10 most empirically rigorous studies. In controlled studies that involved randomization of participants and official reports to measure recidivism, they found an effect size of d=.26 (or a 7% decrease in recidivism beyond traditional criminal justice interventions such as probation or community service). However, in their examination of experimental studies using victim reports as the outcome measure, which is arguably a higher and more accurate estimate of violence recidivism, Feder and Wilson found no effect for batterer intervention programs.
Perhaps one explanation for the general ineffectiveness of batterer intervention programs is the fact that most do not include comprehensive treatment for substance misuse. In fact, men with substance misuse tend to have the poorest batterer intervention program outcomes (e.g., DeMaris & Jackson, 1987; Hamberger & Hastings, 1990; Stuart et al., 2007a). In one study, frequency of male intoxication subsequent to batterer intervention was strongly related to IPV recidivism; batterers who drank almost daily were 16 times more likely to reassault their partners relative to abstinent or seldom drinking men (Jones & Gondolf, 2001). Gondolf (1997) found that 51%-57% of reassaults subsequent to batterer intervention had been preceded within 3 hours by alcohol consumption.
The prevalence of IPV in samples seeking treatment for substance misuse is high. For example, across male inpatient samples, the prevalence of male-to-female IPV in the past year ranged from 58% to 85% (e.g., Bennett et al., 1994; Brown et al., 1998; Chermack et al. 2000; Chermack & Blow, 2002; Gondolf & Foster, 1991; Murphy & O’Farrell, 1994; Murphy et al., 2001). In mixed samples of male inpatients and outpatients, the prevalence of male-to-female violence in the year prior to treatment ranged from 54% to 71% (e.g., Chermack et al., 2000; Chermack et al., 2001; Maiden, 1997; Stith et al., 1991). In outpatient samples of alcoholic males, the prevalence of male-to-female violence in the past year ranged from 54% to 66% (e.g., Murphy & O’Farrell, 1994; Murphy et al., 2001; Stuart et al., 2003c). O’Farrell and Murphy (1995) reported that the prevalence of IPV was 5-6 times higher in their male alcoholic treatment sample relative to a demographically matched nationally representative sample.
Across studies of women substance seeking treatment for substance misuse, research has shown that 50% to 68% perpetrated violence against a partner in the past year, with 25% to 50% reporting severe violence perpetration. For example, Chermack et al. (2001) recruited 126 women in treatment programs for substance misuse and found that over one-half of the women perpetrated violence toward a partner in the past year, and over one-quarter of the women perpetrated severe violence. Chase et al. (2003) recruited a sample of 103 female alcoholics seeking couples-based outpatient alcoholism treatment. They found that in the year prior to treatment, 68% of the women perpetrated at least one act of aggression toward their partners, and 50% engaged in severe violence. Several studies have reported on the victimization of alcoholic women seeking treatment, and consistent with findings from male patients, the prevalence of male-to-female partner violence ranged from 47%-87% (e.g., Chase et al., 2003; Chermack et al., 2000; 2001; Downs, Miller, & Panek, 1993; Haver, 1987).
Data showing an association between alcohol use and violence victimization do not imply that IPV victims are to blame for their victimization (Parks, 2000) but do imply a potential important vulnerability to victimization associated with drinking. It should also be noted that substance misuse may be a consequence of violence victimization; substances may be used as a coping mechanism for the physical and emotional distress associated with IPV (Collins, Kroutil, Roland, & Moore-Gurrera, 1997).
Two studies have examined women’s drinking immediately prior to violence victimization. Parks (2000) found that women drank more, reported feeling more intoxicated, and had a higher blood alcohol level on days in which they were victimized, relative to days in which aggression did not occur. Parks and Fals-Stewart (2004) found that women were 7 times more likely to be victims of sexual aggression and 4 times more likely to be victims of physical aggression on days in which they consumed alcohol relative to days that they did not drink.
Given the high prevalence of IPV in populations with substance misuse and the detrimental consequences of IPV, researchers have investigated the potential impact of treatment for substance misuse on IPV. The review of the literature in this area will be organized in two parts. First, literature is reviewed on the influence of general treatment for substance misuse on violence perpetration and victimization, in which there is no particular emphasis on relationships or IPV in the treatment. In the subsequent section, studies of the impact of behavioral couples therapy on substance use and IPV are reviewed.
Several studies have examined the connection between individual treatment for substance misuse and reductions in IPV. Stuart et al. (2003c) investigated the impact of an individually based intensive partial hospitalization treatment for alcohol dependence on alcohol use, IPV, and psychological aggression among a small sample of heterosexual male patients and their partners. Male patients received 5-6 days of treatment for substance misuse in which there was no direct focus on the intimate relationship. The cognitive-behavioral treatment was multidisciplinary and involved group therapy, daily individual psychotherapy sessions, and daily meetings with a psychiatrist. The program recommended total abstinence from all substances. Psychotherapy groups included functional analysis, understanding irrational thinking, goal setting, managing emotions, assertiveness, spirituality, substance refusal skills, and relapse prevention. Patient and partner assessments were conducted at baseline, 6-month follow-up, and one-year follow-up. Results revealed a decrease in alcohol use in male patients as well as significant reductions in the frequency of male-to-female physical violence and psychological aggression from baseline to 6-month and one-year follow-up. Furthermore, results showed significant decreases in the frequency of female-to-male IPV from baseline through one-year follow-up.
In a similar study, Stuart et al. (2002) examined the potential impact of an intensive outpatient treatment for alcohol dependence on alcohol use, IPV, and psychological aggression among a small sample of heterosexual female alcohol patients and their partners. Patients received the same intensive individual partial hospitalization program described above. Results showed a decrease in alcohol use in female patients as well as declines in the prevalence and frequency of male-to-female partner violence over time. Results also showed decreases in the prevalence and frequency of female-to-male partner violence and psychological aggression.
These studies suggest that IPV is reduced subsequent to individually based alcohol treatment. Other studies of individuals in treatment for substance misuse have reached similar conclusions. For example, Stuart and colleagues (2007b; 2007c) found comparable results to the above studies in larger samples of men and women seeking intensive outpatient treatment for substance misuse. O’Farrell, Fals-Stewart, Murphy, and Murphy (2003) recruited a sample of 301 individuals seeking alcohol treatment. Treatment involved an intake assessment, a physical examination, 8 individual therapy sessions, and 16 group sessions over a 12-week period. They showed that the prevalence of male-to-female IPV decreased significantly, from 56% in the year before treatment (four times the rate of a demographically matched comparison sample) to 25% in the year following individual treatment. They also noted an increased prevalence of IPV for relapsed alcoholics one year after treatment (32%) relative to remitted alcoholics (15%). The prevalence of IPV among remitted alcoholics was virtually identical to the prevalence of IPV in the demographically matched sample.
Fals-Stewart (1998) recruited a sample of 187 male drug misusers seeking individual treatment and found that the prevalence of male-to-female IPV decreased from 60% in the year prior to treatment to 30% in the year following treatment. As in the O’Farrell et al. (2003) study, the prevalence of IPV was substantially higher among relapsed substance misusers relative to remitted substance misusers. In fact, the prevalence of IPV among remitted substance misusers was comparable to the prevalence of IPV among a non-substance misusing control group.
Overall, these studies suggest that IPV perpetration and victimization by male and female patients, and their partners, is reduced subsequent to treatment for substance misuse. Furthermore, research has documented a connection between relapse to substance use and continued relationship aggression.
Although the use of couples therapy with violent couples is controversial, there are compelling reasons for initiating behavioral couples therapy for substance misuse and IPV (O’Farrell & Fals-Stewart, 2006; Stith, Rosen, & McCollom, 2004; Stuart et al., 2007a). First, by addressing relationship problems and substance use, behavioral couples therapy may reduce the substance use and relationship discord that are general risk factors for aggression, thereby decreasing IPV. Couples therapy involves increasing relationship factors conducive to abstinence, including spending more time in couple activities and increasing positive feelings. Second, behavioral couples therapy teaches specific skills to prevent violence. For example, behavioral couples therapy teaches communication and negotiation skills to reduce hostile conflicts that may escalate to violence, as well as problem solving skills to address relationship issues. Therapists typically inquire about IPV and establish goals to prevent its recurrence. Finally, behavioral couples therapy also directly focuses on sobriety by including elements such as a daily sobriety contract, encouraging the use of self help, monitoring urges to drink or use drugs, crisis intervention for substance use, and relapse prevention. Further discussion of these issues can be found in O’Farrell and Fals-Stewart’s (2006) book written on behavioral couples therapy for substance misuse.
Offering couples therapy to individuals with a history of substance misuse and IPV is controversial because there is the concern that this approach may increase the risk of IPV in both partners. Since couples are often advised to speak openly about relationship conflict and distress in treatment, one issue is that either partner (particularly the female partner) may fear that disclosure and addressing relationship problems may escalate the probability of IPV. In general, behavioral couples therapy is only considered if there is a history of low or moderate levels of violence, if both partners independently agree to participate and do not express fear of negative consequences for discussing the relationship and violence, and if both partners commit to avoiding future additional physical aggression. Continuous monitoring of relationship aggression by the therapist takes place at each session and an action plan to stay safe is developed.
O’Farrell, Murphy, Stephan, Fals-Stewart, and Murphy (2004) examined IPV before and after behavioral couples therapy for 303 married or cohabiting male alcoholic patients, and compared their findings to a demographically matched non-alcoholic sample. The treatment involved 10-12 weekly 1 hour conjoint pregroup sessions with each couple, followed by 10 weekly 2-hour couples group sessions. The duration of treatment was 5-6 months. In two studies, male-to-female partner violence was significantly reduced in the two years subsequent to behavioral couples therapy. Moreover, IPV was nearly eliminated for male patients who achieved abstinence from alcohol. In the year prior to behavioral couples therapy, 60% of male alcoholic patients and 64% of their female partners had perpetrated IPV. In the year after behavioral couples therapy, violence decreased significantly among male patients to 24% of the alcoholic sample but remained higher than the comparison group (12%). The prevalence of male-to-female IPV in the year following couples therapy was 12% among male patients whose alcohol problem was in remission, compared to 30% among male patients who relapsed to alcohol. The 12% prevalence of IPV in remitted male alcohol patients was identical to the matched comparison sample. Similar results were found during the second year of follow-up after behavioral couples therapy; specifically, the prevalence of IPV was reduced, with remitted alcoholics evidencing less violence than relapsed alcoholics. Thus, IPV decreased after behavioral couples therapy, and clinically significant violence reductions occurred for patients who ceased drinking after treatment. A series of earlier studies of 88 male alcoholics conducted by O’Farrell and colleagues found comparable results (O’Farrell & Murphy, 1995; O’Farrell et al., 1999; O’Farrell, Murphy, & Neavins, 2000).
In four methodologically rigorous studies conducted by Fals-Stewart and colleagues, behavioral couples therapy reduced intimate partner violence more than individual counseling for substance misuse. Within each of these studies, all patients were assigned the same number of outpatient therapy sessions. Patients were randomized to receive either all individual counseling sessions or half couples therapy and half individual sessions. While participants across all four studies demonstrated a high prevalence of male-to-female IPV in the year prior to treatment, participants demonstrated a substantial reduction in violence in the year subsequent to receiving treatment. In addition, participants receiving behavioral couples therapy exhibited greater reductions in IPV, demonstrated more abstinence from alcohol and drugs, and reported greater relationship satisfaction than participants who received individual counseling.
In a sample of 80 male patients who misused drugs, particularly cocaine and heroin, Fals-Stewart, Kashdan, O’Farrell, and Birchler (2002) found a comparable prevalence of male to female IPV prior to treatment (above 40% for both the group receiving behavioral couples therapy and the group receiving individual counseling only); however, a smaller proportion (18%) of men receiving behavioral couples treatment engaged in IPV in the year after treatment, relative to men receiving only individual treatment (43%). Similarly, in a sample of 138 female alcoholic patients, Fals-Stewart, Birchler, and Kelley (2006) found that women randomized to behavioral couples therapy evidenced significantly reduced male-to-female and female-to-male IPV relative to women randomized to individual counseling. Studies examining other populations revealed virtually identical results, including a sample of 195 male alcoholic patients (Birchler & Fals-Stewart, 2003) and a sample of 215 cocaine dependent patients (Fals-Stewart, O’Farrell, & Birchler, 2003).
Studies examining the influence of treatment for substance misuse on IPV carry important treatment implications. First, given the high prevalence of IPV in populations of individuals in treatment for substance misuse, it is crucial that substance misuse treatment providers assess IPV in all of their patients involved in intimate relationships. IPV is associated with serious negative consequences, and clients often do not spontaneously report IPV to their individual or couples therapists (Holzworth-Munroe & Stuart, 1994; O’Farrell & Fals-Stewart, 2006; O’Leary, Vivian, & Malone, 1992). Thus, we recommend that treatment providers conduct thorough assessments of IPV perpetrated by both partners. As part of this assessment, it is important to directly query clients about physical, verbal, psychological, and sexual aggression, as well as the dangerousness and possible lethality of the relationship aggression. We believe that such assessments should become a standard part of clinical practice.
In populations of individuals in committed relationships seeking treatment for substance misuse, we strongly recommend participation in behavioral couples counseling. Numerous studies, including tightly controlled randomized clinical trials, have documented the efficacy of behavioral couples therapy. These studies have shown that behavioral couples therapy elicits superior IPV outcomes relative to individual therapy. In addition to the superior outcomes on IPV, studies comparing behavioral couples therapy with individual-based treatment show a consistent pattern of more abstinence from alcohol and drugs, fewer substance-related problems, happier relationships, and lower risk of relationship separation for individuals who received behavioral couples therapy (e.g., O’Farrell & Fals-Stewart, 2006).
To reiterate, behavioral couples therapy should only be considered if there is a history of low to moderate levels of IPV, if both partners independently agree to participate and do not express fear of negative consequences for doing so, and if both partners firmly commit to nonviolence. In cases in which these conditions are not met, other approaches may be warranted. First, among these options is individual treatment for substance use; research indicates that such treatment is likely to have a positive impact on relationships and IPV, albeit substantially less than behavioral couples therapy. In light of the evidence showing that individuals who relapse to substance misuse engage in more IPV than individuals who are remitted, more attention should be paid to relapse prevention. Indeed, such a focus would likely yield positive outcomes across a variety of domains.
Although there is some potential for individual treatment for substance misuse to reduce IPV, other alternatives could also be considered. For example, Rychtarik and McGillicuddy (2005) recruited 171 women who were partners of male alcoholics and compared the effects of coping skills training for women and Al-Anon Facilitation therapy. Coping skills focused on addressing how the women could deal with their male partner’s drinking and to avoid being victimized by IPV. Specific attention was paid to coping with situations in which the male partner was intoxicated (e.g., leaving the situation, not arguing with him). Relative to Al-Anon Facilitation therapy, coping skills training resulted in less IPV victimization, particularly for women whose partners drank greater amounts. This research suggests that there may be effective treatment options to further reduce IPV victimization for partners of those who misuse substances.
Interventions for violence-related problems could be integrated into treatment for substance misuse, or provided as adjuncts to substance use treatment, in an effort to further reduce IPV (e.g., Chermack & Blow, 2002). For example, Schumacher, Coffey, Stasiewicz, Leonard, and Fals-Stewart (2007) have begun a small randomized clinical trial in which men in treatment for substance misuse with a past-year history of perpetrating IPV are randomized to an additional adjunct brief intervention for IPV or assessment alone. Their preliminary results are promising. Another viable approach to preventing the occurrence and escalation of IPV would be to target all individuals attending a substance misuse treatment program regardless of individual risk (i.e., selective intervention; see Temple, Stuart, & O’Farrell, in this issue).
Given the high prevalence of substance misuse in batterers, the finding that batterer intervention programs may have limited efficacy in preventing further IPV (Babcock et al., 2004; Feder & Wilson, 2005), and that men arrested for IPV who misuse substances have the poorest batterer intervention outcomes (DeMaris & Jackson, 1987; Hamberger & Hastings, 1990; Stuart, 2005; Stuart et al., 2007a), it may be wise to offer adjunct or integrated interventions for substance misuse (Stuart, 2005; Stuart et al., 2007a). Jones and Gondolf (2001) found that the probability of violence recidivism subsequent to participation in a batterer intervention program was reduced by 30% - 40% if the individual obtained treatment for substance misuse. Though linking batterer programs and substance use treatment has proven challenging (e.g., Collins et al., 1997), it is logical to offer treatment for substance misuse to the large proportion of batterers who need it.
Although the body of research on the effects of individually-based and couples-based interventions for substance misuse and IPV is considerable, there is much work yet to be done. First, as described in Temple et al.(in this issue), virtually no work has been conducted examining the prevention of IPV in substance misusers. In addition, the work integrating anger management into substance misuse programs, and work involving adjunct brief interventions for IPV in substance use treatment, is in its infancy. Extensive efforts to research these areas should be undertaken.
Given the positive outcomes to date, we also recommend continued use of randomized clinical trial methodologies for behavioral couples therapy in other populations. For example, the research to date conducted on the impact of behavioral couples therapy has primarily involved couples in which only one partner, but not both, engaged in addictive behaviors. Additional studies are needed to examine the potential benefits of couples approaches when both partners misuse substances. Indeed, preliminary work with these samples has demonstrated promising results (Schumm, O’Farrell, & Murphy, 2007). Similarly, researchers have recently begun to document the efficacy of behavioral couples therapy for substance misuse among gay and lesbian couples (Fals-Stewart, O’Farrell, & Birchler, 2007), and are beginning to tailor couples therapy for cigarette smoking cessation (LaChance, 2007). In addition, the efficacy of couples therapy for IPV and substance use could be examined using other modalities. For example, the effects of behavioral couples therapy could be conducted in groups of couples, rather than with individual couples, to increase efficiency and cost effectiveness.
In populations of individuals who have been arrested for IPV and are involved in court mandated violence intervention programs, research should be conducted on both integrated and adjunct treatments for substance misuse. We believe that many men and women arrested for domestic violence would benefit significantly from obtaining concurrent treatment for substance misuse. As an example, we are in the process of conducting a randomized clinical trial in which men and women with alcohol misuse who are arrested for violence receive either the standard batterer intervention program or the standard batterer intervention program combined with a brief, motivational intervention for hazardous drinking. Preliminary data indicate that participants receiving the brief alcohol treatment evidence better substance use and partner violence outcomes relative to those in standard care. However, this research is still in its early stages, and there is much more work to be done. Future research could also investigate the effects of behavioral couples therapy subsequent to completion of batterer intervention. Ideally, the couple would then be free of violence, which would allow them to work on improving their relationship and preventing the recurrence of violence.
Finally, more research should be conducted on treatment failures. Whether individuals in treatment for substance misuse receive individual therapy, behavioral couples therapy, batterer intervention programs, anger management, coping skills training, Twelve-Step Facilitation therapy, and/or motivational interventions, relapse and IPV are likely to remain significant problems. Research on how best to match substance misusers with IPV to treatments, or combinations of treatment approaches, would be informative.
Dr. Stuart is supported, in part, by grants R01AA016315 and R01AA014193 from the National Institute on Alcohol Abuse and Alcoholism. Dr. O’Farrell is supported by grants from the National Institute on Alcohol Abuse and Alcoholism (R01AA14700, R01AA12834), the National Instutte on Drug Abuse (R01DA15156), and by the Department of Veterans Affairs. Dr. Temple is supported by a research career development award (K12HD052023: Building Interdisciplinary Research Careers in Women’s Health Program - BIRCWH) from the National Institute of Allergy and Infectious Diseases (NIAID), the National Institute of Child Health and Human Development (NICHD), and the Office of the Director (OD), National Institutes of Health.
The following definitions are adapted from the Centers for Disease Control and Prevention (CDC) website: www.cdc.gov.
Gregory L. Stuart, PhD, is an associate professor in the department of Psychology at the University of Tennessee - Knoxville. He was formerly an associate professor of Psychiatry and Human Behavior in the Alpert Medical School of Brown University at Butler Hospital. His research focuses primarily on the comorbidity of intimate partner violence and substance misuse. He is particularly interested in interventions that address both substance use and relationship aggression.
Timothy J. O’Farrell, Ph.D., ABPP is Professor of Psychology in the Harvard Medical School Department of Psychiatry at the VA Boston Healthcare System where he directs the Families and Addiction Program and the Counseling for Alcoholics’ Marriages (CALM) Project. His clinical and research interests focus primarily on couple and family therapy in alcoholism and drug abuse treatment and various aspects of substance abusers’ family relationships including partner violence, child functioning, and sexual adjustment. The title of his most recent book is Behavioral Couples Therapy for Alcoholism and Drug Abuse (2006).
Jeff R. Temple, PhD, is an assistant professor in the department of Obstetrics & Gynecology at the University of Texas Medical Branch at Galveston. He has been conducting research on intimate partner violence (IPV) since he began his graduate studies at the University of North Texas and later as a predoctoral intern and postdoctoral research fellow at Brown University. His research interests include violence prevention and intervention, along with the mental and physical health effects of IPV and on factors that influence this relationship (e.g., substance misuse, social support).
Gregory L. Stuart, University of Tennessee - Knoxville.
Timothy J. O’Farrell, Harvard Medical School Department of Psychiatry at the VA Boston Healthcare System.
Jeff R. Temple, University of Texas Medical Branch at Galveston.