Intimate partner violence (IPV) is a serious, yet preventable international public health concern. IPV is the self-reported experience of physical and/or sexual violence by a current or former partner since the age of 15 years old (World Health Organization [WHO] 2013); stalking and psychological aggression were later added as types of IPV (Breiding et al. 2015). In their effort to understand the global impact that IPV encompasses, WHO (2013) obtained global and regional estimates of physical and/or sexual IPV based on data extracted from 79 countries and 2 territories. The global lifetime prevalence of IPV among ever-partnered women was found to be 30 %. The highest prevalence was in the African, Southeast Asia, and Eastern Mediterranean WHO regions where approximately 37 % of ever-partnered women reported IPV at some point in their lives.

Data regarding male victims of IPV may be significantly underreported due to the perception of the public and law enforcement. Men who report IPV may be seen as cowardly, feel embarrassed, and/or fear being laughed at or scorned (Shuler 2010). Furthermore, few men report their abuse to law enforcement due to the fear of disbelief and support services offered (Allen-Collinson 2009). As a result, male victims do not freely admit being a victim of IPV at the hands of females and, therefore, do not seek professional intervention (Barber 2008). In their review of the previous 10 years of IPV research, Desmarais et al. (2012) identified that approximately one in five men has experienced physical violence in an intimate relationship. This serious international public health concern warrants further exploration into our prevention, assessment, and treatment efforts.

The Complexity of Intimate Partner Violence

Violence is a multifaceted construct. Megargee (1982) described four domains that influence criminal violence: instigation, inhibition, habit strength, and situation. The instigation domain is the sum of the internal influences such as cognitions, motivations, and feelings that incline a person to behave violently, whereas inhibition is the sum of internal influences that decrease the likeliness that a person would behave violently. Habit strength refers to the static history of violent and nonviolent behavior. The situation domain consists of the external factors that impact violence. Meloy (1988) identified forensic criteria for affective and predatory modes of violence that appear to stem from the instigation and situation domains of Megargee (1982). Affective violence is an instinctual, defensive, and reactive form of violence to a perceived threat. The goal of affective violence is threat reduction. Predatory violence is a planned or purposeful form of violence where there is no perceived threat; it is attack-focused, and the goals are variable (Meloy 1988). The psychological determinants and the modes of violence have framed the field of violence risk assessment for individual violence, interpersonal violence, and collective violence.

IPV is one component under the broader topic of family violence. When focusing on a public health concern, assessment, or treatment intervention, behavioral scientists turn to and complete research from a variety of perspectives. Theoretical explanations for family violence generally fall into the following areas: macrotheory, microtheory, learning theory, intrapersonal difference theories, systems theory, and multidimensional theories (Barnett et al. 2011).

Research has identified numerous types of IPV perpetrators. In their work on differentiating types of IPV, Kelly and Johnson (2008) have identified the following four typologies of IPV: coercive controlling violence, violent resistance, situational couple violence, and separation-instigated violence. Coercive controlling violence occurs when the perpetrator uses physical and emotional violence to intimidate, dominate, control, and coerce the victim. The violent resistance typology refers to instances when the victim may resist the coercive controlling violence with their own retaliatory violence. The situational couple violence typology is the most common type of physical aggression committed by cohabitating and married men and women. This type of violence in IPV has a variety of causes and stems from the couple’s inability to resolve conflict, manage anger, and effectively communicate; these situations or arguments can escalate to physical violence. Lastly, the separation-instigated violence typology indicates that separation/divorce is the catalyst for violent acts. The acts in this typology are unexpected and uncharacteristic and there is no prior history of violence.

To provide more information on the complexity of IPV, experts in the field of violence risk assessment have a general consensus based on their comprehensive reviews on risk factors for IPV. Upon reviews of hundreds of studies touching on risk issues, perpetrator risk factors for IPV include: past antisocial behavior/attitudes, attitudes supportive of violence, history of violence in intimate relationships, personality disorders, substance abuse, prior threatening or stalking behavior, homicidal or suicidal ideation, sexual proprietariness, recent employment or financial struggles, recent relationship problems, and the minimization or denial of violent acts (Kropp and Cook 2014). Despite the identification of these targeted risk factors, they are not typically targeted into IPV intervention models.

Overview of the Duluth Model

The Duluth Model was developed by Domestic Abuse Intervention Programs (DAIP). This model is the most commonly used intervention in the United States and Canada for men who are court-sanctioned to treatment for a conviction of a domestic assault type of offense (Corvo et al. 2009). The Duluth Model is rooted in feminist and sociocultural concepts of domination and control where IPV is used as a means for men to exhibit power and establish control over their female partners (Pence and Paymar 1993). The prominent tool of the Duluth Model is the Power and Control Wheel which delineates how men use male privilege, emotional and economic abuse, violence, intimidation, and isolation to control women. The Duluth Model focuses on the coordination of community responses, which set out to empower and protect the survivors of domestic violence while holding the perpetrators accountable (Pence and Paymar 1993; Mankowski et al. 2002). Although the format of the Duluth Model is educational, it does incorporate cognitive-behavioral techniques.

Support for the Duluth Model

Although dated, the Duluth Model does have some merit. In their study that utilized measures to review dominant models of partner aggression, Burge et al. (2015) obtained 200 participants; 79 (the larger of the groups) reported IPV in their relationship that was predictable, controllable, and stemmed from the power and control structure of the Duluth Model. In a long term follow-up study with batterers within the Duluth model program, Herman et al. (2014) utilized the Relationship Beliefs Scale (RBS). The authors found that 63 % of the participants agreed with the statement that women provoke IPV by using bad judgment or by provoking the man’s anger; 45 % believed that men have the natural right to be in charge of relationships. There was a statistically significant difference between the pretest and posttest measures on the RBS; this difference is indicative of the men in this study experiencing a re-education of the previously mentioned maladaptive thoughts. These supportive studies further reinforce that the aspects of power and control should not be ignored in cases of IPV.

Controversy over the Duluth Model

In their meta-analysis using IPV-related keywords in MEDLINE and PsychINFO, Stover et al. (2009) identified IPV interventions (batterer, couple, victim, and child witness) that met the following criteria: experimental study, sample size of at least 20 subjects per group, and the outcome variable measures recidivism or violence severity. The authors found mandatory arrest, the Duluth Model, and group CBT as the most common interventions for batterers. In their review, all three interventions have minimal ability to break the cycle of violence with most studies demonstrating little to no impact above mandatory arrest. Furthermore, based on victim’s reports, the authors identified that one in three cases will recidivate within 6 months when perpetrators receive these interventions.

The Duluth Model has received increasing amounts of criticism. Critics of the Duluth Model challenge the intervention as a relevant science as the creators of the model openly identify its inception by a “small group of activists in the battered women’s movement” (Pence and Paymar 1993, p. xiii) as opposed to qualified health professionals. Another critique stems from the model being designed for use by paraprofessionals. Given the unique intrapersonal characteristics and interpersonal dynamics of the perpetrator and the victim in IPV cases, rehabilitation efforts should be administered by qualified health professionals. These concerns appear to be valid and warrants further consideration. Effective interventions generally link to a diagnosis and have evidence supporting its use. This lack of scientific evidence in the model’s creation coupled with evidence indicating that the model is ineffective (e.g., Schrock and Padavic 2007) is problematic seeing that the Duluth Model is the predominant intervention in the United States and Canada for men who are court-sanctioned (Corvo et al. 2009).

Pender (2012) provided an analysis of the Duluth Model according to the Association for Specialists in Group Work (ASGW) Best Practice Guidelines (2008). The ASGW is a division of the American Counseling Association; a well-respected organization and leader in understanding and evaluating group work. In her analysis, Pender (2012) identifies that the Duluth Model does not require facilitators to have a graduate/professional degree or continuing education requirements (paraprofessionals). Therefore, significant questions surrounding competency as well as the accountability of the facilitators to a governing body is in question. Pender (2012) was unable to identify program evaluation criteria, therefore it does not appear to meet the evaluation criteria in the ASGW Best Practice Standards. From a diversity standpoint, the Duluth Model is solely used for men who batter women with no consideration for women who batter men or for same-sex couples. Currently, the program is only appropriate for Caucasian, African American, Native American, and Latino populations. The Duluth Model continues to remain limited in terms of treating diverse populations. Despite these significant limitations, Pender (2012) found the coordinated community response (utilizing all departments/organizations in IPV cases) and the collaborative nature of the program as strengths of the model.

Corvo et al. (2009) completed a thorough review of ethical principles from the field of psychology, social work, counseling, and marriage and family therapy when compared to the Duluth Model. They cited numerous ethical violations in each respective discipline. Given the relatively poor outcomes of the Duluth Model, informed consent requires the practitioner to inform the client of the known poor outcomes of these interventions; it is unclear if this is provided at the time of informed consent. Also, practitioners are to use interventions based on established scientific knowledge, newly supported clinical research, support through rigorous scientific study (e.g. meta-analysis), or evidence-based practices. The Duluth Model was developed by activists, not professionals, and is designed to be used by paraprofessionals. This is problematic given that the Duluth Model is the predominant model for intervention, yet it is not grounded in science. Lastly, the use of the term “batterer” is derogatory and inaccurate. Practitioners must refrain from using derogatory language regarding their clients by putting people first, not their disability.

Multicultural psychology is concerned with the effect that culture (an external factor) has on internal processes (cognitions and emotions). In order to be culturally competent, the practitioner must be able to identify the numerous diversity variables (race, ethnicity, gender, etc.), have an appreciation and acceptance of these differences (multiculturalism), and possess an awareness of personal belief systems. As previously mentioned, the Duluth Model is very narrow in its scope. Intervention is solely for males of Caucasian, African American, Native American, and Latino descent who perpetrate from heterosexual intimate relationships. This does not consider female perpetrators, same-sex relationships, or perpetrators of other races. Given that the Duluth Model was developed by activists, lacks a scientific base, does not require continuing education, has no governing body for practitioners, and is designed to be practiced by paraprofessionals, there is minimal evidence that culturally competent services are being provided to the clients via this model.

The last controversial issue is the Duluth Model’s narrow use of the very diverse field of the behavioral sciences. The Duluth Model is based on feminist theory and the sociocultural concept of power and control; the other disciplines in the behavioral sciences and theoretical explanations for family violence are not included in its understanding of IPV. Furthermore, this model is very outdated and important research updates outside of feminist theory are not included. As previously mentioned, the four domains of violence (instigation, inhibition, habit strength, situation) proposed by Megargee (1982) as well as the types of violence (affective vs. predatory) identified by Meloy (1988) are important when conceptualizing violence. Lastly, having IPV-specific typologies (coercive controlling, violent resistance, situational couple, separation-instigated) and risk factors for violence helps practitioners to understand the various forms of IPV and factors associated. However, the Duluth Model forgoes this information and favors the simplistic explanation of violence as the patriarchal view of the male needing to establish power and control. This narrow focus of IPV does a disservice to perpetrators and victims.

Analysis and Recommendations

The goal of this writing is not to denounce the Duluth Model as there is some support and merit to the work that the DAIP has done in this area. Some of the critique that the model has received may be unwarranted and untrue. For example, the DAIP has addressed accusations that they promote a shame-based intervention, do not account for anger as a causal factor, are anti-marriage, and that they ignore psychological problems. The issue of power and control in relationships cannot be ignored. Since power and control is the primary component of the Duluth Model, there do appear to be elements that can be clinically useful. Despite some support, the information presented does support that a change is necessary.

As previously mentioned, IPV perpetration is a very complex issue that cannot be primarily explained by patriarchal power and control. The evidence surrounding the diversity and complexity of IPV highlights the limited ability of the Duluth Model. The primary components were the model not considering same-sex relationships, having a limited ability to be extended to a variety of races, and focused only on male perpetrators.

The Duluth Model exhibits a minimal understanding of violence as well as IPV. In terms of violence, this model only appears to address the two components of the instigation domain by Megargee (1982); power and control. Inhibition, habit strength, and situation do not appear to be addressed. However, the DAIP, in conjunction with the City of Duluth have created the Crossroads Program. This program is designed for women who use illegal violence towards their male partner who document that the male has a history of abusing them (affective violence). This does exhibit an understanding of affective violence, however, the focus is still narrow as it does not involve males who are reactively violent. From the family violence perspective, the Duluth Model only appears to exhibit one theory (feminism) within microtheory; it does not include the other theories of family violence. The Duluth Model does address two of the four IPV typologies provided by Kelly and Johnson (2008): coercive controlling and violent resistance violence. The coercive controlling typology is identical to power and control whereas the Crossroads Program exhibits an understanding of violent resistant violence. However, the situational couple violence is the most prominent typology which the model does not address. Lastly, the plethora of risk factors for violence and IPV has been established and well-documented; these factors are not considered in the model.

Based on the information presented, including the meta-analyses, current interventions for perpetrators of IPV are ineffective; this includes the Duluth Model as the predominant model. In order to address this issue, the establishment of a conclusive theory that includes the known origins (typologies/models) of IPV coupled with the flexibility to address the variances among the various diversity variables (culture, gender, race, sexuality, etc.) is necessary. This conclusive theory will then be able to drive prevention, assessment, and intervention practices and procedures that can hopefully improve the treatment outcomes. Ultimately, the development of an evidence-based practice and incorporation within the criminal justice and legal system is necessary. Furthermore, qualified health professionals, as opposed to paraprofessionals must be providing the service given the complex nature of IPV cases.