Download Effectiveness of Systemic Treatment for Intimate Partner Violence: A Review and more Study notes Literature in PDF only on Docsity!
SYSTEMIC PERSPECTIVES ON INTIMATE PARTNER
VIOLENCE TREATMENT
Sandra M. Stith
Kansas State University
Eric E. McCollum
Virginia Tech
Yvonne Amanor-Boadu
Kansas State University
with Douglas Smith
Texas Tech University
This article reviews changes in the research literature on intimate partner violence (IPV) since our earlier review (Stith, Rosen, & McCollum, 2003). A rationale for systemic treatment of IPV has emerged from research that has continued to document the limited effectiveness of single-gender treatment approaches for offenders and that has identified subtypes of abusive relationships, including situational couple violence, which often includes the reciprocal use of violence. Consistent findings from the available outcome research have demonstrated that for carefully screened couples who choose to stay together, systemic interventions decrease incidences of IPV and decrease the risk factors for IPV with no increase in risk. Implications for research and treatment are offered.
This article reviews changes in the research literature on intimate partner violence (IPV) since our earlier review (Stith, Rosen, & McCollum, 2003). A number of changes have occurred in the way we view IPV as well as in our knowledge about existing and developing treatment approaches. In this article, we describe the research that has led to these changes, as we seek to understand and deal with what remains a serious and costly social problem. As there is a very limited amount of research on violence in same-sex relationships, and no research on the effectiveness of treatment approaches for same-sex couples, the research that is reviewed here is reflective of that limitation.
An Evolving Understanding of IPV As a result of research in the field, our understanding of IPV is beginning to broaden and change. Traditionally, IPV has been seen through a feminist paradigm and understood to be the expression of men’s power over women, occurring in intimate heterosexual relationships, and supported by a patriarchal culture. Violence was considered a male phenomenon with women either remaining solely victims or assaulting their male partners in self-defense. Violence was also seen as the primary problem, with coexisting issues often seen as distractions that helped men evade responsibility for their violence. The responsibility for violence was unilater- ally men’s, while the costs were unilaterally women’s and the focus of intervention was to end violence specifically against women. Evidence for this view of IPV came from studies using criminal justice and shelter-seeking populations and showed considerable gender asymmetry (i.e., many more men than women are arrested, and many more women than men seek shelter in domestic violence victim shelters). However, this view was challenged as research using community samples began to be
Sandra M. Stith, PhD, Kansas State University; Eric E. McCollum, PhD, Virginia Tech; Yvonne Amanor- Boadu, PhD, Kansas State University; Douglas Smith, PhD, Texas Tech University. Address correspondence to Sandra M. Stith, Family Studies and Human Services, 101 Campus Creek Complex, Kansas State University, Manhattan, Kansas 66506; E-mail: sstith@ksu.edu
Journal of Marital and Family Therapy doi: 10.1111/j.1752-0606.2011.00245.x January 2012, Vol. 38, No. 1, 220–
220 JOURNAL OF MARITAL AND FAMILY THERAPY January 2012
conducted. Although research examining arrested offenders or victims seeking shelter continues to show dramatic gender asymmetry, community-based studies find that IPV perpetration and victimization may be more gender symmetrical than we previously thought, with participants reporting male-perpetrated, female-perpetrated, and reciprocally perpetrated violence. Whitaker, Haileyesus, Swahn, and Saltzman (2007), for instance, analyzed data on 11,370 U.S. adults aged 18 to 28 from the 2001 National Longitudinal Study of Adolescent Health and found almost 24% of all relationships had some violence. Interestingly, this study found that half of those relationships were reciprocally violent; that is, both partners assaulted each other. Fur- thermore, in those relationships where the violence was unilateral, women were the perpetrators in more than 70% of the cases. Methods used to assess violence, including sampling strategies, influence prevalence rates, yet it is also becoming clearer that both men and women perpetrate IPV. Family therapists need to focus our efforts not only on ending violence against women but also on ending all forms of violence in relationships. Not only has our understanding of the high prevalence of male-perpetrated, female- perpetrated, and reciprocal violence increased in the past decade, but we are also beginning to understand more about the impact of violence on both men and women. Archer (2000), in a meta-analysis, found that while more women than men reported perpetrating violence, 62% of those injured by an intimate partner were women. Tjaden (2000) found that 26.4% of male IPV victims and 32.6% of the female victims reported that their partner threatened to harm or kill them, although female victims were twice as likely to report being fearful of bodily injury or death than male victims (44.7% vs. 19.6%). While the assumption might be that male-perpetrated violence results in more actual danger, Whitaker et al. (2007) found, in fact, that reciprocal IPV was associated with greater injury than was nonreciprocal IPV, regardless of the gender of the perpetrator. Only a few studies have compared mental health outcomes of IPV victimization for men and women. Afifi et al. (2009) used data available from the U.S. National Comorbidity Survey Replication (NCS-R) study to examine the psychological effects of male and female IPV victim- ization. They reported that IPV was associated with poor mental health outcomes for both men and women, although women experienced a wider range of problems than did men. Male victims of IPV were more likely than men in nonviolent relationships to experience externaliz- ing disorders, including disruptive behavior disorders and substance use disorders. Female vic- tims of IPV were more likely than women in nonviolent relationships to experience both externalizing and internalizing disorders (anxiety disorders) and suicidal ideation. Other research has supported these findings. Using a subsample of 7,395 married and cohabiting heterosexual couples drawn from Wave 1 of the National Survey of Families and Households, Anderson (2004) found that although IPV is associated with negative health consequences for both women and men, it is associated with significantly more depression and substance abuse for women. While the experience of victimization may have differing impacts on men than on women, it is important to recognize that both men and women are injured and are psychologi- cally impacted by victimization. As family therapists become more aware that IPV is not nearly as gender asymmetrical as we once thought, the importance of providing treatment for both partners in an ongoing committed relationship becomes more apparent. Our understanding of IPV has been further expanded as we have also begun to examine different typologies of violence—both types of violent offenders and types of violent relation- ships. One of the most widely researched typologies of violent relationships was developed by Johnson and Ferraro (2000; Johnson, 2006), who identified four types of violent heterosexual couples: those experiencing ‘‘situational couple violence,’’ ‘‘intimate terrorism,’’ ‘‘violent resis- tance,’’ and ‘‘mutual violent control.’’ Intimate terrorism generally involves unilateral violence and includes a high level of coercive control. In contrast, situational violence is more likely to be bilateral and involves conflict over a particular issue. Violent resistance involves vio- lence that is enacted to resist intimate terrorism and may have the primary motive of wanting to protect oneself, or be the result of an expression of anger or resistance to a controlling partner. Mutual violent control includes two equally coercive partners engaged in a struggle for control of the relationship. Situational violence is hypothesized to be the most prevalent type of relationship violence, particularly within samples from the general population, and in
January 2012 JOURNAL OF MARITAL AND FAMILY THERAPY 221
IPV for 15 months. Men participating in IPV treatment were 8 times more likely and men in alcohol treatment were 11 times more likely to be physically aggressive toward their partner on days they consumed alcohol. More recently, Wupperman et al. (2009) reported that most female partners of men in IPV treatment in their study reported not only being violent with their partners but also that they were as likely as their partners to use substances the week prior to their partners’ participation in IPV treatment. The body of research is clear that IPV and substance abuse are directly related. Fals-Stewart and Kennedy (2005), in fact, argue that the evidence supports a causal relationship between substance abuse and IPV and that substance abuse must be addressed in IPV treatment attempts. Later, in this article, we discuss research on evidence-based treatment approaches for the co-occurring substance abuse.
EFFECTIVENESS OF TRADITIONAL TREATMENT
Accompanying the traditional male perpetrator or female victim paradigm for understand- ing IPV in heterosexual relationships was a specific approach to intervention. Men were adjudi- cated through the court system and mandated to attend all-male batterers’ intervention programs (BIPs), while women were offered voluntary victim support services. The BIPs were designed to challenge men’s use of male power and teach new, egalitarian ways of relating. Despite the widespread use of this model, a growing body of research has called its efficacy into question.
Treatment of Male Offenders Research into the effectiveness of BIPs has continued since the publication of our last review. In addition to effectiveness studies of individual programs, meta-analyses of the effec- tiveness research in this area have been published, as well as review articles that have identified the challenges involved in such studies. Two meta-analyses have made important contributions to our knowledge of the outcome of BIP intervention. Babcock, Green, and Robie (2004) con- ducted a meta-analysis of 17 quasi-experimental and 5 experimental studies investigating the treatment effectives of BIPs for partner violent men. Treatment outcome was measured in terms of victim report of re-assault and ⁄ or police reports of re-assault. For studies measuring recidi- vism by police report, those with an experimental design had an average treatment effect of d = .12, and those with a quasi-experimental design had an average treatment effect of d = .23, showing a significant but small impact on recidivism for both types of studies. For the studies measuring recidivism by partner report, those with an experimental design had an aver- age treatment effect of d = .09, representing a nonsignificant impact on recidivism, and those with a quasi-experimental design had an average treatment effect of d = .34, again representing a significant but small impact on recidivism. Babcock et al. (2004) also examined differences in treatment effects based on type of treatment, comparing studies using the Duluth model, to cognitive-behavioral therapy (CBT) groups and ‘‘other’’ interventions and found no significant differences in effect sizes based on treatment type. While there is some question about what these small effect sizes actually mean for women who have been assaulted by an intimate part- ner, Babcock et al. (2004) note that, using the most conservative result, the treatment effect based on partner report in experimental studies (d = .09), treatment is responsible for approximately one tenth of standard deviation improvement in recidivism. In other words, a man who is arrested, sanctioned by the court, and treated has a 40% chance of remaining non- violent versus a 35% chance of remaining nonviolent for a man who is arrested and sanctioned but not treated. Feder and Wilson (2005) also conducted a meta-analysis of BIP outcome studies, using more rigorous inclusion criteria than did Babcock et al. (2004), resulting in a sample of four experimental and six quasi-experimental studies. Again, treatment outcome was measured in terms of partner or police report of re-assault. Based on police reports, the average effect size for experimental studies was d = .26, representing a significant but small effect—a reduction in recidivism from 20% to 13%. Quasi-experimental-designed studies were broken into two groups and analyzed separately. Those comparing men who were mandated to treatment versus those not mandated to treatment had an average effect size of d = ).14, a nonsignificant effect.
January 2012 JOURNAL OF MARITAL AND FAMILY THERAPY 223
Those comparing men who completed mandated treatment versus those who were rejected from treatment, who never attended, or who dropped out had an average effect size of d = .97, rep- resenting a significant and large treatment effect. However, the authors express reservations about this finding, noting that treatment completers may be significantly different from those who are rejected from treatment or who fail to attend or drop out; for instance, they may be more highly motivated or more fearful of criminal justice sanctions, and, thus, the treatment effect may be confounded by these factors. For studies using partner report of re-assault, Feder and Wilson (2005) report a nonsignificant average effect size of near zero for experimental stud- ies and a small and negative nonsignificant average effect size for quasi-experimental studies using a no-treatment control group. Thus, this meta-analysis again indicates that treatment effects for BIPs are small to nonexistent, particularly in the stronger studies—those using an experimental design and ⁄ or partner reports of re-assault. While the findings from these two meta-analyses are discouraging, there are a number of challenges in investigating the impact of treatment on re-assault rates that may contribute to the small effect sizes found in these studies (Babcock et al., 2007; Eckhardt, Murphy, Black, & Suhr, 2006; Gondolf, 2004). For instance, meta-analyses may still be difficult when the number of available studies, particularly those that utilize an experimental design, is limited. Using an experimental design is also challenging in this field when a ‘‘no-treatment’’ control group also often includes monitoring by the criminal justice system, and dropout rates of 40%–60% often mean that the treatment group contains many individuals who either never received treatment or received a very low dose. Additionally, while partner reports of re-assault may be preferable to official police reports, given that repeat incidents may not always be reported to the police, outcome studies frequently find very low response rates from partners. The length of follow-up period also presents challenges, as re-assault rates may be higher immediately after completing a program but decline over time, yet maintaining contact with partners for an extended period is problematic. Finally, outcome studies may demonstrate small effect sizes because of chal- lenges inherent in the BIP treatment itself. For instance, court-mandated individuals may not be highly motivated to change, and group treatment approaches, while demonstrating some effi- cacy for internalizing behaviors, have not been shown to be as effective for externalizing behav- iors such as IPV (Babcock et al., 2007). Additionally, there may be components of the BIPs that are less effective than others for all types of offenders. These challenges have led to another focus within BIP research: efforts to find components or interventions that will increase motiva- tion, reduce dropout, and reduce recidivism. One new area of study has been the application of motivational interviewing (MI) interven- tions (Miller & Rollnick, 2002) to work with IPV offenders in efforts to increase their engage- ment with and attendance at group treatment programs. Taft, Murphy, Elliott, and Morrel (2001) investigated the effectiveness of motivational enhancement techniques including tele- phone calls and handwritten letters to group participants before the commencement of group treatment and immediately following any absences from group. They found significant effects on group attendance compared to a group that did not receive the intervention. Those in the treatment retention group attended roughly 10% more sessions than did the control group, a small-to-medium effect size (d = .35). Additionally, only 15% of men in the treatment reten- tion group dropped out, compared to 30% in the control condition. Finally, there was a signifi- cant treatment by race interaction effect, with 42% of minority individuals dropping out of the control group and only 10% of minority individuals dropping out of the treatment retention group. Effects of attendance on both partner reports of abusive behaviors and criminal justice data of recidivism indicated that higher attendance was associated with less recidivism, lower reported physical assault and injury at posttreatment, and lower injury at 6-month follow-up. While Taft et al. (2001) included their MI intervention throughout the course of BIP treat- ment, another MI strategy has been to administer brief interventions prior to entry into the BIP group. Results of this approach are mixed. Kistenmacher and Weiss (2008) conducted a randomized controlled trial of a two-session MI pregroup intervention with 33 men court- mandated to BIP treatment to determine the effects of a brief MI intervention on offenders’ readiness for change. Those receiving the MI intervention showed a significantly greater pre-to-post increase in readiness to change and a significantly greater pre-to-post decrease
224 JOURNAL OF MARITAL AND FAMILY THERAPY January 2012
participants, including information about mood disorders and PTSD, parenting information, and safety planning. Fifty-eight percent of participants dropped out of the program. Dowd et al. (2005) found no differences between dropouts and treatment completers in demographic variables, in the history of violence in families of origin, in previous relationships, or in current relationships or in past mental health treatment. They did find, however, that women who entered the program voluntarily had fewer pretreatment arrests, were more likely to drop out after intake, and had lower completion rates overall than did court-mandated women. Clearly, as more women are appearing for treatment of IPV, we need to know more both about women’s use of violence in intimate relationships and about what constitutes effective interven- tion for women.
SYSTEMIC TREATMENT OF IPV
Research on the effectiveness of systemic treatment or intervention for IPV is growing, but questions remain. In this section of the article, we discuss one major research program (O’Farrell & Fals-Stewart, 2002), four completed projects, and two projects in the beginning phase of research designed to treat or prevent IPV using systemic interventions. While some of these programs indicate that they are designed to prevent IPV and others to treat IPV, none of these programs are designed to provide primary prevention to the general population. Most of these programs are designed to prevent low-level or situational violence from escalat- ing to more severe violence. Therefore, we review them as a group. The programs reviewed are listed in Table 1. We begin this section, however, with a discussion of screening for IPV in sys- temic therapies, as accurate identification of IPV and risk assessment are the cornerstones of safe and effective treatment.
MFTs and Assessment of IPV As many couples choose to remain together after experiencing violence, they often present for family therapy. In fact, studies have demonstrated that between 36% and 58% of couples who seek regular outpatient treatment have experienced male-to-female physical assault in the past year and between 37% and 57% have experienced female-to-male physical assault (Jose & O’Leary, 2009). As a result of the high level of IPV in couples coming to therapy, several authors have offered guidelines for universal screening for IPV when working with couples (Bograd, 1999; Stith et al., 2003). General guidelines suggest screening all couples using individ- ual interviews with both partners, and multimodal assessments (e.g., written questionnaires and verbal interviews). Careful screening and ongoing monitoring are the basis for determining whether conjoint therapy can proceed safely. In fact, assessment of physical violence has been included in the American Association for Marriage and Family Therapy (AAMFT) Core Competencies recommended for accredited training programs. Despite this recognition of the need for assessment, there has been very little published research to document how, or if, assessment is being conducted by marital and family thera- pists (MFTs). Todahl, Linville, Chou, and Maher-Cosenza (2008) conducted a qualitative study of MFT interns’ experiences of universal screening for IPV with 22 (17 women and 5 men) MFT interns who had an average of 400 client contact hours. All of the interns had partici- pated in an IPV class that addressed universal screening guidelines. Todahl et al. (2008) found wide variations in screening practices. Eleven interns reported routine screening, whereas others reported waiting for ‘‘red flags’’ to appear before assessing for violence. Thirteen routinely separated couples to screen for violence, whereas six separated couples only rarely, and only three interns stated that they used a combination of verbal questions and written questionnaires to assess for violence. In identifying barriers to universal screening, 14 interns expressed concern for victims’ safety, fearing that asking about violence would put victims at further risk or re-traumatize them. Others reported that a lack of confidence in their ability to both assess for and treat IPV was a barrier for them. Schacht, Dimidjian, George, and Berns (2009), using a sample of 620 practicing MFTs randomly selected from the AAMFT membership list, found similar variability in assessment procedures and limited adherence to universal screening guidelines. Just over half (53.2%) of
226 JOURNAL OF MARITAL AND FAMILY THERAPY January 2012
Table 1 Couples Treatment for Domestic Violence Experimental DesignAuthor, year, andname of program
Comparisonconditions
Sample size
Outcome
Outcomemeasure
Follow-upperiod
LaTaillade et al.(2006)
10 sessions cognitive-behavioralcouples therapy
17 couples
Significant decrease inpsychological aggression forboth groups; no significantpre–post difference in physicalaggression for either group
CTS
4 months
Couples AbusePreventionProgram
10 sessions systemic couplestreatment at family therapycenter
21 couples
Woodin andO’Leary(2010)
2-hr assessment and 2-hrmotivational feedback session
25 collegestudent couples
Motivational sessions led tosignificantly greaterreduction in physicalaggression
CTS2 (eitherpartner’sreport of violence)
9 months
2-hr assessment and minimal,nonmotivational feedback session
25 collegestudent couples
Fals-Stewart andClinton-Sherrod(2009)
Behavioral couples therapyfor married or cohabitingmale substance-abusingpatients and their partners(12 conjoint sessions, 20 individual12-step sessions)
103 couples
BCT led to significantly lessmale-to-female violencecompared to IBT
Male-to-femalepercentage ofdays with anyviolence
12 months
Individual-based treatment formale substance-abusing partneronly (32 individual 12-step sessions)
104 couples
Mills (2008)
Circles of Peace (weeklyconferences with offender,victim, family, and communitymembers)
Total
N
152 domesticviolence offenders
Clients in Circles of Peacehad significantly fewerarrests overall
Arrest records
24 months
Circles of Peace
Batterer intervention program
No significant differencein domestic violencearrests
January 2012 JOURNAL OF MARITAL AND FAMILY THERAPY 227
participants in this study reported that they screened all couples that they saw the previous year for violence, 42.3% reported screening some couples, and 4.5% reported screening no couples. The latter group was excluded from further analysis. Of those who screened for IPV, 37.2% reported always interviewing partners separately, 54.9% reported interviewing some partners separately, and 7.9% reported never interviewing partners separately. Finally, 78.9% reported that they did not use a written self-report instrument during the screening process, and of those who did use a written instrument, only 7.5% reported using standardized, behaviorally specific questionnaires designed to measure violence. Only 3.5% of participants reported using all of these procedures described above that constitute appropriate screening for IPV. In addition, Schacht and colleagues’ (2009) participants varied widely in the criteria they used to determine whether conjoint therapy is appropriate when violence is detected. Fifty-two percent reported that they considered the overall prognosis of the relationship, 42% considered the victim’s level of fear and safety in the relationship, 40.5% considered the severity, fre- quency, and duration of abuse, 30.2% considered psychopathology of either victim or perpetra- tor, and 23.1% considered whether either the victim or perpetrator was engaged in other therapy services. Of the least endorsed items, only 5.6% considered the potential effect of cou- ple therapy on current levels of violence, 4.8% considered injury or lethality of past violence, 1.7% considered the use or presence of weapons, and 1.4% considered the perpetrator’s history of violence in other relationships. Although based on limited research, it appears that the MFT field falls short of the level of universal screening that has been recommended in the literature. This is a concern given that careful assessment is generally a foundation on which safe use of the systemic models we will describe below is built.
Behavioral Couples Treatment The research group led by O’Farrell and Fals-Stewart (2002) has contributed both clinical intervention development and testing and more basic research to our understanding and treat- ment of IPV in couples with concurrent substance abuse disorders. In this article, we will focus on the outcome studies of their clinical intervention: behavioral couples treatment (BCT). Behavioral couples treatment is a dyadic intervention used to treat adults with substance abuse disorders. The couple is seen conjointly, and the non-substance-abusing partner is enlisted as a support for the substance-abusing partner’s sobriety. The couple is helped to negotiate a Sobriety Contract that includes a daily Sobriety Trust discussion in which the substance-abusing partner reaffirms his or her intent not to use alcohol or drugs that day. The non-substance-abusing partner provides positive support for that intention. Some patients also participate in self-help groups or use medications to support abstinence and take those in the presence of their partner. Discussions, self-help attendance, medication use, and relapse are recorded on a calendar by the couple and brought to sessions for further discussion, support of success by the therapist, and troubleshooting in cases of relapse. In addition to support for sobriety, BCT also includes skill training modules that increase positive interactions and teach communication skills in the service of managing conflict better. Beginning with small-scale pilot studies in the 1970s, the program has grown to include several large, federally funded random- ized controlled clinical trials with a strong body of evidence for the efficacy of BCT in reducing both substance abuse and IPV (Fals-Stewart, O’Farrell, Birchler, Cordova, & Kelley, 2005). Behavioral couples treatment began as a successful treatment specifically for alcoholic men and their female partners. In the course of their work on substance abuse, the BCT group began to be interested in the effect of BCT on IPV. There is consistent evidence that both problems co-occur with regularity (see Fals-Stewart, Klostermann, & Clinton-Sherrod, 2009, for a summary of this literature), and the BCT group began to investigate the impact of BCT on couple violence. Fals-Stewart et al. (2005) report that, for alcoholic men, BCT results in dramatic reductions in IPV after treatment. In two studies (O’Farrell, Murphy, Hoover, Fals- Stewart, & Murphy, 2004; O’Farrell, Van Hutton, & Murphy, 1999), the BCT group found that IPV was significantly reduced for male alcoholics following treatment. In the larger 2004 study, the prevalence rate for IPV in a matched nonalcoholic comparison sample was 12% in the year prior to assessment, while 60% of male alcoholic patients had been violent to their female partners during the same time period. Following treatment, BCT reduced the rate of
January 2012 JOURNAL OF MARITAL AND FAMILY THERAPY 229
IPV in the alcoholic sample to 24% overall—a significant reduction although still higher than the comparison group. Among those men who were no longer drinking, the rate was reduced to 12%—equal to the nonalcoholic group. Among men who relapsed in the year after treat- ment, the rate rose to 30%. Thus, for alcoholic men, BCT resulted in reduced violence, with the largest reductions being associated with abstinence from alcohol. Among men abusing substances other than alcohol, Fals-Stewart, Kashdan, O’Farrell, and Birchler (2002) found that nearly 50% of couples reported IPV against a female partner in the year prior to treatment. After treatment, 17% of couples receiving BCT reported male violence, while 42% of couples in which the male partner participated in an equally intensive individual treatment program reported violence. These findings are based on female partners’ reports. Since their 2005 review article, the BCT group has continued to investigate the effect of BCT on IPV among substance-abusing patients. Schumm, O’Farrell, Murphy, and Fals-Stewart (2009) examined the impact of BCT on partner violence for two years following treatment in a sample of married or cohabiting women who sought treatment for alcoholism with their male partners. In the year prior to treatment, both the women and their male partners had higher prevalence and frequency of aggression than did a matched nonalcoholic sample. For couples who received BCT, there were significant decreases in both the first and second year following treatment for both alcohol-abusing women and their male partners on all aggression measures except that male-perpetrated severe violence was not significantly reduced in the first year after BCT. As with male alcoholic patients, abstinence was associated with better violence outcomes. Couples in which the female partner was abstinent were not significantly different from the nonalcoholic sample, except that the male partners of abstinent female participants had greater prevalence and frequency of verbal aggression in the year following treatment compared to those in the matched sample. In a subsequent study, Fals-Stewart and Clinton-Sherrod (2009) randomly assigned their sample of 207 substance-abusing men and their female partners to either BCT treatment or an individual-based treatment (IBT) for the male substance-abusing partner, to determine the impact of participation in BCT on the relationship between substance use and occurrences of IPV. In this version of BCT, the couples’ sessions included specific attention to violence pre- vention and safety. For instance, couples made a verbal agreement not to engage in any angry touching, and if their partners relapsed, women were coached not to engage in any kind of conflict resolution discussion with them. The individual treatment condition (IBT) also included 32 sessions, composed of 1-hour individual-based sessions following a 12-step facilita- tion model. In examining differences between groups on substance abuse, there were no differences in percentage of days abstinent (PDA) between those in the BCT or the IBT at pretreatment or posttreatment. At 12-month follow-up, however, those in the BCT group had significantly higher PDAs than those in the IBT. A similar pattern was seen in rates of violence, with no differences between groups at pre- or posttreatment, but at 12-month follow-up, those who had received BCT had lower rates of any violence and lower rates of severe violence than those who had received IBT. On days of no substance use, the likelihood of nonsevere and severe violence did not differ for those assigned to the BCT or the IBT group. The likelihood of non- severe and severe violence increased significantly on days the man used substances among men who received IBT. However, on days of substance use, the likelihood of IPV was lower for men who had received BCT compared to those receiving IBT. While it is not yet clear what the specific mechanism of action is that gives BCT the advantage in preventing violence on days of drinking, it does appear that BCT fares better than IBT when risk is higher. Overall, the work of the BCT group has provided strong support for systemic interventions to address substance abuse and also to reduce IPV. Individual treatment of substance abuse did not have nearly the impact on IPV as did systemic treatment. Although this is an encourag- ing outcome, it must be noted that BCT is typically administered to couples in which only one partner is using substances (Fals-Stewart, O’Farrell, & Birchler, 2004). Given the high associa- tion between substance abuse and IPV for both partners in a couple, this may leave out a significant number of couples seeking treatment.
230 JOURNAL OF MARITAL AND FAMILY THERAPY January 2012
a problem, be willing to work toward an abuse-free relationship, be committed to staying together, and feel safe participating in conjoint treatment. The CAPP treatment protocol has been pilot tested in comparison with treatment as usual (TU) at a University of Maryland family and couple therapy clinic (LaTaillade et al., 2006). In this pilot study, couples who met the inclusion criteria and consented to participate in the study (47.5% of those who were eligible) were randomly assigned to CAPP (n = 17 couples) or TU (n = 21 couples). The TU condition was composed of individual couple therapy based on a variety of systems-informed therapy models, depending upon the framework used by the thera- pist assigned to the case. Treatment in the CAPP condition followed a structured format with each session focused on a particular content area. In both treatment conditions, couples attended ten 90-minute sessions, clients were asked to sign no-violence contracts, and treatment was focused on reducing systemic patterns that lead to IPV. In the CAPP condition, the first session included an overview of the program and discus- sion of the relationship history, completion of a no-violence contract, and identification of a written set of goals for therapy, with an understanding that the primary goal of CAPP is for the couple to have an abuse-free relationship. For homework, couples were asked to review their goals, revise them if they wish, and bring them to the following session. In the second session, co-therapists refined treatment goals, educated partners about cognitive-behavior constructs, taught strategies for anger management, and provided education about the conse- quences of constructive versus destructive forms of communication. For homework, partners were asked to practice the anger management strategies. In sessions 3 and 4, therapists taught expressive and listening skills and had partners practice these skills in session and for home- work, along with practicing anger management skills for homework. Sessions 5 through 7 provided instruction and practice skills for resolving conflict without abuse, and partners were coached in combining communication and problem-solving skills. Finally, in sessions 8 through 10, the communication and problem-solving skills were supplemented with relationship recovery
Table 2
DVFCT Significant Changes for Males and Females
Single couple n = 17
Multicouple group n = 28
Comparison n = 9
Physical aggression (partner report)
M F M F
Psychological aggression (partner report)
M M F
Marital conflict M F Marital satisfaction F M F Constructive communication F M F Destructive communication M Partner pursues, respondent distances
F M
Respondent pursues, partner distances
M
Anger M Anxiety M Respondent differentiation F M F M Partner differentiation M F
Note. DVFCT = domestic violence–focused couples treatment; M = males; F = females.
232 JOURNAL OF MARITAL AND FAMILY THERAPY January 2012
and enhancement strategies to increase the proportion of positive activities and sharing, develop greater mutual support, increase affection and intimacy, and increase partners’ ability to work as a team in setting and working toward goals. Additionally, in session 10, therapists summa- rized the couple’s progress toward their initial treatment goals and addressed relapse prevention through the identification of skills that had been learned and ways to maintain progress that had been achieved. Research on CAPP found largely similar outcomes compared to TU. Relationship satisfac- tion is increased and psychological aggression decreased on at least some measures in both conditions, while there were no differences in physical aggression likely due to the low frequency of physical aggression in the sample to begin with. Based on coded communication measures, CAPP produced less negative communication for both men and women, while there were no such changes in TU. Men in the CAPP condition tended toward more positive communication, while there were no changes in positive communication in TU.
Circles of Peace Beginning with principles of restorative justice, Mills (2008) developed a systemic interven- tion called Circles of Peace (CP) as an alternative to traditional BIP treatment. Circles of Peace involves conferences between victims and offenders that also include roles for family members and friends who provide support and care to each individual involved in a crime (Grauwiler, 2004). A CP is made up of a Circle Keeper, typically a community member trained in working with IPV, the offender (termed ‘‘the applicant’’), and the victim (‘‘the participant’’), if she or he chooses to participate. In addition, extended family members, friends, and ⁄ or community mem- bers may become involved to support the individuals, and one person is designated as the ‘‘safety monitor,’’ who monitors the family between conferences. A CP involves the use of an intake assessment that includes a safety screening to ensure that it is safe for the victim to par- ticipate, and an ‘‘Initial Social Compact’’ is a document signed by the offender promising not to be violent and to participate in any other treatments that might be necessary. Circles further involve the use of a ‘‘talking piece,’’ an object identified by the family that must held by the speaker when talking, and begin with the rules of no violence, no blaming, and a focus on acknowledgment, understanding, responsibility, and healing. Mills received funding from the National Science Foundation to study the effectiveness of CP compared to a traditional BIP treatment in Nogales, Arizona. In this study, 152 court adju- dicated cases were randomly assigned to CP or BIP, and recidivism data in the form of subse- quent arrest records were obtained for 24 months following treatment for all those assigned to either treatment, whether or not they completed treatment. Groups were comparable at base- line, and over half of the CP offenders had victims who agreed to participate in the CP pro- gram. Preliminary results, which have not yet been published (Mills & Barocas, 2009), indicate that CP offenders had significantly fewer subsequent overall arrests in the 24 months following treatment than did BIP offenders, including fewer arrests for IPV; however, the differences between groups in subsequent IPV arrests were not statistically significant. Mills and her research team are currently in the process of conducting a follow-up study that will include a larger sample size and two phases of research, the first comprising a comparison between BIP treatment and BIP plus CP treatment and the second comprising a comparison between BIP, BIP plus CP, and BIP plus a couple’s treatment program based on a Couples Conflict Group.
Motivational Interviewing Woodin and O’Leary (2010) have reported on the effectiveness of a targeted brief moti- vational interviewing (MI; Miller & Rollnick, 2002) intervention to prevent IPV in high-risk heterosexual dating couples. In this study, 50 college students and their partners underwent a 2-hr assessment session and then were randomly assigned to receive either the MI feedback condition or the minimal feedback condition. Couples had to have been dating for at least 3 months, with no history of marriage or cohabitation and at least one act of male-to-female physical aggression reported by either partner on the Revised Conflict Tactics Scale (CTS2; Straus et al., 1996). At the assessment session, partners independently filled out questionnaires assessing partner aggression, problem alcohol use, acceptance of partner aggression,
January 2012 JOURNAL OF MARITAL AND FAMILY THERAPY 233
intimacy, coping with stress and depression, and including information about the importance of fathers and healthy marriage. The 22 sessions address five content areas: Managing Conflict; Managing Stress; Fathers, Marriage, and Parenting; Creating Shared Meaning; and Maintaining Intimacy. Importantly, the project plans to look not only at outcome but also at the mechanisms specifically responsible for any observed decrease in IPV. Preliminary results from data collected before the program began and at posttest with 115 couples indicated that the program was suc- cessful in strengthening relationships and decreasing conflict (Bradley, Friend, & Gottman, in press); however, no significant differences in levels of violence between treatment and control groups or between pre- and posttests for the treatment group were found in the preliminary analyses (R. C. Bradley, personal communication, November 2010). The research team will be continuing to collect and analyze follow-up data at 6 months and 12 months following comple- tion of the program. In a project funded by the Centers for Disease Control, Heyman and Slep are evaluating Couple Care for Parents (CCP), an intervention for new parents under age 30 aimed at preventing the development or escalation of IPV. The intervention was originally developed and tested in Australia. Modified for a North American clientele and augmented with material to specifically address IPV, CCP is a psychoeducational intervention aimed at decreasing stress, improving com- munication and parenting skills, and maintaining couple intimacy after the birth of a child. In a randomized controlled trial testing Couple CARE for Parents, men and women with a newborn were assessed shortly after the baby was born and randomly assigned to either the Couple CARE for Parents of Newborns program, or a waitlist control that would get Couple CARE for Parents of Toddlers after the 24-month assessments were over. Preliminary analyses indicate that while men in the control group report significant increases from birth to 8 months in the amount of physical and psychological aggression they receive from their partners, men in the treatment group show no significant increases. Furthermore, men in the treatment group show significant decreases in received aggression from birth to 15 months. Preliminary analyses also indicate that women in the control group report significant decreases in relationship satisfaction from birth to 8 months, but women in the treatment group report no significant change in satisfaction. How- ever, women in both groups did not see a change in received physical or psychological aggression (D. Mitnick, personal communication, December 23, 2010). Follow-up data collection is ongoing.
LIMITATIONS OF THE RESEARCH TO DATE
Despite the encouraging findings from extant studies of systemic interventions, as well as the fact that funded projects are in process, issues still remain that need to be addressed. No work, to date, has addressed the processes involved in changing violent relationships, although the Gottman project plans to do so. Thus, we do not know what aspects of these interventions lead to change and what aspects of these interventions may be unnecessary. Researchers have made efforts to assess fidelity to treatment within their protocols; however, little of this work has been published. While one of the concerns expressed by those who deliver batterer intervention programs is that conjoint treatment approaches are more expensive than male-only approaches, no research has been conducted on the cost-effectiveness of systemic treatment approaches, especially in comparison with batterer intervention approaches. We also do not know much about for whom these interventions are most effective and ⁄ or for whom they might be ineffective or even dangerous. Most current projects have fairly stringent exclu- sion criteria and do not allow highly violent couples or couples in which intimate terrorism is occurring to participate in the treatment. In addition to a lack of diversity in types of violence, the research to date has not evaluated the use of systemic interventions with clients from diverse cultural backgrounds nor have researchers examined their use with same-sex couples. Furthermore, we lack effectiveness research studies to examine the use of systemic interventions in ‘‘real-world settings.’’ We need future research that addresses factors leading to change, which more clearly examines who does and does not change as a result of these interventions, and research in settings more likely to represent the settings in which most clinicians practice. Another issue deserving attention is the widening gap between research and practice when it comes to systemic interventions. Despite evidence that systemic interventions can be useful in
January 2012 JOURNAL OF MARITAL AND FAMILY THERAPY 235
deceasing violence and improving couple relationships, the standard clinical practice continues to be separate gender group interventions for perpetrators using pro-feminist or cognitive- behavioral approaches (Saunders, 2008). Regardless of the studies calling the effectiveness of separate gender intervention into question, these programs are institutionalized in state stan- dards for IPV intervention across the country. Forty-five states currently have standards for IPV intervention, and of those, 95% mandate a curriculum based on power and control with or without attention to social psychological issues such as skill deficits and faulty modeling in the perpetrator’s family of origin (Maiuro & Eberle, 2008). More to the point, 68% of state standards explicitly prohibit conjoint couples’ treatment during the primary phase of IPV inter- vention, while the remaining 32% remain silent on the issue or limit the circumstance in which couples’ sessions can be held. Not only do such standards limit the ability of couples to access a treatment that can be helpful to them if they remain together in the wake of violent acts—or if they want to separate safely yet co-parent their children—they also pose difficulties for researchers who wish to test conjoint approaches in community agencies with the populations to which they are likely to be delivered. The existence of state standards that prohibit the use of systemic interventions may also contribute to the lack of research that is accretive in this field. Other than the work conducted by Fals-Stewart et al. (which is conducted within substance abuse programs), there are no pro- grams of research studies in this area that build on each other nor have researchers other than the model developers themselves been involved in evaluating systemic interventions. Further- more, while state standards may place constraints around the types of intervention programs that may be used for arrested offenders, funding limitations also exist in this area, as the focus of much federal funding has narrowed to the two areas of DSM diagnoses and prevention programs.
CONCLUSION
Intimate partner violence continues to be a significant social problem with major gaps in our understanding of how best to intervene, including understanding which specific factors con- tribute to reductions in physical aggression for different types of perpetrators. Current accepted treatments appear not to be living up to the promise they once held, and newer approaches remain controversial and often rejected by those on the frontlines. As a violent crime occurring in an attachment relationship, IPV garners the attention of social institutions with widely diver- gent approaches. The judicial system’s reliance on sanction and rehabilitation collides with the mental health system’s approach of understanding and reconciliation, for instance. It is also dif- ficult to know how to categorize these acts when we see severely injured victims unwilling to leave their abusers or couples who declare unequivocally their love for one another yet psycho- logically attack each other in extreme ways. Systemic therapists have much to contribute to how society addresses IPV, but we cannot do this work in isolation. We must remain part of a coor- dinated community approach. Perhaps the promising treatments we are writing about 10 years from now will include models that integrate what, at this point, remain opposing viewpoints.
A CLINICIAN RESPONDS
Douglas Smith
The authors provided a detailed review of the current state of the relationship violence lit- erature that suggests a number of important clinical implications. Despite increased research and clinical attention, researchers continue to report that intimate partner violence occurs in an alarming number of relationships. In clinical populations, the rates of IPV may approach 50% (O’Leary, Vivian, & Malone, 1992). The clear implication is that all marriage and family thera- pists will work with couples and families who are affected by violence. Therefore, it is essential that MFTs possess basic knowledge about assessment and intervention with IPV. Perhaps the most significant implication of the study findings presented in the literature review is that clini- cians must reevaluate traditionally accepted schemas about relationship violence.
236 JOURNAL OF MARITAL AND FAMILY THERAPY January 2012
the profession. However, perhaps in response to legitimate concerns raised by the feminist cri- tique of family therapy, the field has resisted the application of systemic concepts to clinical intervention with IPV. It is encouraging that emerging research is providing support for a more systemic conceptualization of IPV and support for systemic intervention with IPV. As we move forward, the interests of our clients will be best served by developing efficacious interventions for IPV that focus on eliminating relationship violence while honoring the lessons from our past by attending to issues of power and safety.
REFERENCES
Afifi, T. O., MacMillan, H., Cox, B. J., Asmundson, G. J. G., Stein, M. B., & Sareen, J. (2009). Mental health correlates of intimate partner violence in marital relationships in a nationally representative sample of males and females. Journal of Interpersonal Violence, 24 , 1398–1417. Anderson, K. L. (2004). Perpetrator or victim? Relationships between intimate partner violence and well-being. Journal of Marriage and Family, 64 , 851–863. Archer, J. (2000). Sex differences in aggression between heterosexual partners: A meta-analytic review. Psychologi- cal Bulletin, 126 , 651–680. Babcock, J. C. (2003). Toward a typology of abusive women: Differences between partner-only and generally violent women in the use of violence. Psychology of Women Quarterly, 27 , 153–161. Babcock, J. C., Canady, B. E., Graham, K., & Schart, L. (2007). The evolution of battering interventions: From the dark ages into the scientific ages. In J. Hamel & T. L. Nicholls (Eds.), Family interventions in domestic violence: A handbook of gender-inclusive theory and treatment (pp. 215–244). New York: Springer. Babcock, J. C., Green, C. E., & Robie, C. (2004). Does batterers’ treatment work? A meta-analytic review of domestic violence treatment. Clinical Psychology Review, 23 , 1023–1053. Bograd, M. (1999). Battering and couples therapy: Universal screening and selection of treatment modality. Jour- nal of Marital and Family Therapy, 25 , 291–312. Bradley, R. C., Friend, D. J., & Gottman, J. M. (in press). Supporting healthy relationships in low-income, vio- lent couples: Reducing conflict and strengthening relationship skills and satisfaction. Journal of Couple and Relationship Therapy. Buehlman, K. T., Gottman, J. M., & Katz, L. F. (1992). How a couple views their past predicts their future: Predicting divorce from an oral history interview. Journal of Family Psychology, 5 , 295–318. Carney, M. M., & Buttell, F. P. (2004). A multidimensional evaluation of a treatment program for female batter- ers: A pilot study. Research on Social Work Practice, 14 , 249–258. Dowd, L., Leisring, P. A., & Rosenbaum, A. (2005). Partner aggressive women: Characteristics and treatment attrition. Violence and Victims, 20 , 219–233. Eckhardt, C. I., Murphy, C., Black, D., & Suhr, L. (2006). Intervention programs for perpetrators of intimate partner violence: Conclusions from a clinical research perspective. Public Health Reports, 121 , 369–
Fals-Stewart, W. (2003). The occurrence of partner physical aggression on days of alcohol consumption: A longi- tudinal diary study. Journal of Consulting and Clinical Psychology, 71 , 41–52. Fals-Stewart, W., & Clinton-Sherrod, M. (2009). Treating intimate partner violence among substance-abusing dyads: The effect of couples therapy. Professional Psychology: Research and Practice, 40 , 257–263. Fals-Stewart, W., Kashdan, T. B., O’Farrell, T. J., & Birchler, G. R. (2002). Behavioral couples therapy for drug- abusing patients: Effects on partner violence. Journal of Substance Abuse Treatment, 22 , 87–96. Fals-Stewart, W., & Kennedy, C. (2005). Addressing intimate partner violence in substance-abuse treatment. Journal of Substance Abuse Treatment, 29 , 5–17. Fals-Stewart, W., Klostermann, K., & Clinton-Sherrod, M. (2009). Substance abuse and intimate partner violence. In K. D. O’Leary & E. M. Woodin (Eds.), Psychological and physical aggression in couples: Causes and interventions (pp. 251–269). Washington, DC: American Psychological Association. Fals-Stewart, W., O’Farrell, T. J., & Birchler, G. R. (2004). Behavioral couples therapy for substance abuse: Rationale, methods, and findings. Science and Practice Perspectives, 2 , 30–41. Fals-Stewart, W., O’Farrell, T., Birchler, G., Cordova, J., & Kelley, M. (2005). Behavioral couples therapy for alcoholism and drug abuse: Where we’ve been, where we are and where we’re going. Journal of Cognitive Psychotherapy, 19 , 229–246. Feder, L., & Wilson, D. (2005). A meta-analytic review of court-mandated batterer intervention programs: Can courts affect abusers’ behavior? Journal of Experimental Criminology, 1 , 239–262. Gondolf, E. (2004). Evaluating batterer counseling programs: A difficult task showing some effects and implica- tions. Aggression and Violent Behavior, 9 , 605–631. Gottman, J. (1994). Why marriages succeed or fail. New York: Simon and Schuster.
238 JOURNAL OF MARITAL AND FAMILY THERAPY January 2012
Gottman, J., & Silver, N. (1999). The seven principles for making marriage work. New York: Crown. Grauwiler, P. (2004). Moving beyond the criminal justice paradigm: A radical restorative justice approach to inti- mate abuse. Journal of Sociology and Social Welfare, 31 , 49–69. Harway, M., & Hansen, M. (1993). Therapist perceptions of family violence. In M. Harway & M. Hansen (Eds.), Battering and family therapy: A feminist perspective (pp. 42–53). Newbury Park, CA: Sage. Jacobson, N., & Gottman, J. (1998). When men batter women: New insights into ending abusive relationships. New York: Simon and Schuster. Johnson, M. P. (2006). Conflict and control: Gender symmetry and asymmetry in domestic violence. Violence Against Women, 12 , 1003–1018. Johnson, M. P., & Ferraro, K. J. (2000). Research on domestic violence in the 1990s: Making distinctions. Jour- nal of Marriage and the Family, 62 , 948–963. Jose, A., & O’Leary, K. D. (2009). Prevalence of partner aggression in representative and clinic samples. In K. D. O’Leary & E. M. Woodin (Eds.), Psychological and physical aggression in couples: Causes and interventions (pp. 15–35). Washington, DC: American Psychological Association. Kernsmith, P. (2005). Exerting power or striking back: A gendered comparison of motivations for domestic vio- lence perpetration. Violence and Victims, 20 , 173–185. Kistenmacher, B., & Weiss, R. L. (2008). Motivational interviewing as a mechanism for change in men who bat- ter: A randomized controlled trial. Violence and Victims, 23 , 558–570. LaTaillade, J., Epstein, N. B., & Werlinich, C. A. (2006). Conjoint treatment of intimate partner violence: A cog- nitive behavioral approach. Journal of Cognitive Psychotherapy, 20 , 393–410. Maiuro, R. D., & Eberle, J. A. (2008). State standards for domestic violence perpetrator treatment: Current sta- tus, trends, and recommendations. Violence and Victims, 23 , 133–155. Miller, W. R., & Rollnick, S. (2002). Motivational interviewing: Preparing people for change (2nd ed.). New York: Guilford Press. Mills, L. G. (2008). Violent partners: A breakthrough plan for ending the cycle of abuse. New York: Basic Books. Mills, L., & Barocas, B. (2009). A comparison study of batterer intervention and restorative justice programs for domestic violence offenders. Unpublished report. Musser, P. (2008). Motivational interviewing as a pregroup intervention for partner-violent men. Violence and Victims, 23 , 539–557. O’Farrell, T. J., & Fals-Stewart, W. (2002). Behavioral couples and family therapy for substance abusers. Current Psychiatry Reports, 4 , 371–376. O’Farrell, T. J., Murphy, C. M., Hoover, S. A., Fals-Stewart, W., & Murphy, M. (2004). Domestic violence before and after couples-based alcoholism treatment: The role of treatment involvement and abstinence. Journal of Clinical and Consulting Psychology, 72 , 202–217. O’Farrell, T. J., Van Hutton, V., & Murphy, C. M. (1999). Domestic violence before and after alcoholism treat- ment: A two-year longitudinal study. Journal of Studies on Alcohol, 60 , 317–321. O’Leary, K. D., Vivian, D., & Malone, J. (1992). Assessment of physical aggression against women in marriage: The need for multimodal assessment. Behavioral Assessment, 14 , 5–14. Saunders, D. G. (2008). Group interventions for men who batter: A summary of program descriptions and research. Violence and Victims, 23 , 156–172. Schacht, R. L., Dimidjian, S., George, W. H., & Berns, S. B. (2009). Domestic violence assessment procedures among couple therapists. Journal of Marital and Family Therapy, 35 , 47–59. Schumm, J., O’Farrell, T., Murphy, C., & Fals-Stewart, W. (2009). Partner violence before and after couples- based alcoholism treatment for female alcoholic patients. Journal of Consulting and Clinical Psychology, 77 , 1136–1146. de Shazer, S. (1985). Keys to solution in brief therapy. New York: Norton. de Shazer, S., Dolan, Y., Korman, H., Trepper, T., McCollum, E. E., & Berg, I. K. (2007). More than miracles: The state of the art of solution-focused brief therapy. Binghamton, NY: Haworth. Simpson, L. E., Doss, B. D., Wheeler, J., & Christensen, A. (2007). Relationship violence among couples seeking therapy: Common couple violence or battering? Journal of Marital and Family Therapy, 33 , 270–283. Smith Stover, C., Meadows, A. L., & Kaufman, J. (2009). Interventions for intimate partner violence: Review and implications for evidence-based practice. Professional Psychology: Research and Practice, 40 , 223–233. Spanier, G. (1976). Measuring dyadic adjustment: New scales for assessing the quality of marriage and similar dyads. Journal of Marriage and Family, 38 , 15–28. Stith, S. M., & McCollum, E. E. (2009). Couples treatment for physical and psychological aggression. In D. K. O’Leary & E. M. Woodin (Eds.), Understanding psychological and physical aggression in couples: Existing evidence and clinical implications (pp. 233–250). Washington, DC: American Psychological Association. Stith, S. M., McCollum, E. E., & Rosen, K. H. (2011). Couples treatment for domestic violence: Finding safe solu- tions. Washington, DC: American Psychological Association.
January 2012 JOURNAL OF MARITAL AND FAMILY THERAPY 239
