Compendium 2000 on Effective Correctional Programming

CHAPTER 13

Treatment of Family Violence in Correctional Settings

LYNN STEWART, JIM HILL, and JANICE CRIPPS¹

---

This chapter provides a brief review of issues related to the treatment of spousal violence. Specifically, it will focus on the assessment and treatment of male offenders who have abused their women partners. We begin with a brief description of the range of abusive behaviours and statistics on the prevalence of spousal violence in Canada followed by theory derived models of intervention. The discussion then focuses on therapeutic factors related to appropriate treatment targets, therapist and offender characteristics and other responsivity issues. A brief section on treatment issues unique to Aboriginal offenders is included. Assessment issues are reviewed, with a focus on pre and post treatment assessment. The final section presents a review of treatment outcome literature and discusses some of the problems associated with program evaluation in that area.

THE SCOPE AND NATURE OF THE PROBLEM

Official statistics provide under estimates of the actual incidence of family violence. Many cases go unreported by victims for a number of reasons such as their fear of reprisal, fear of loosing children to Social Services and lack of alternative accommodation. Some victims may simply not realize that spousal assault is illegal. Nevertheless, the official rates are high enough to illustrate that spousal violence is a serious problem in Canada. In 1997, there were 22,254 reported spousal violence incidents (Fitzgerald, 1999). Of these, 88% involved women victims. Although the general surveys have been restricted to research on heterosexual couples, several smaller studies have indicated that the prevalence of abusive relationships in gay and lesbian relationships is also high (Lockhart et al., 1994; Waldner-Haugrud & Gratch, 1997).

There is evidence that offenders may have particularly high rates of perpetrating family violence. A file review indicated that 27% of federal offenders who had been in a marital relationship were violent to a woman partner at least once (Robinson & Taylor, 1995). More recently, a survey of federal offenders assessed at intake indicated that 40% were found to have some file evidence of abuse against a partner that required a more in depth assessment. Of these, 82% were assessed as being high or medium risk to continue to be abusive (Kropp, 1998). Risk markers for family violence such as: criminal histories, previous histories of violence, alcohol abuse, low education and high unemployment, violence in the family of origin and personality disorder (Dutton & Hart, 1993) are characteristics of a high percentage of federal offenders. For example, 80% of federal offenders have committed a violent offence, 61% have histories of alcohol abuse, 70% test below the grade level and 71% have unstable job histories (Johnson & Grant, 1998).

Abusive behaviours can be categorized into three major groupings: physical, sexual and psychological/emotional. Physical abuse in a relationship is defined as any form of physical violence perpetrated on a partner. In 1997, approximately 74% of spousal assault incidents were classified as common assault, 14% of incidents were classified as either assault causing bodily harm or aggravated assault (Statistics Canada, 1999). Unlike other forms of spousal abuse, in cases of assault causing bodily harm or aggravated assault, men are more likely to be the victims, possibly because women resort to weapons to compete against the greater physical strength of their male partners. However, homicide rates in Canada indicate that male violence is more likely to lead to the death of a woman. From 1978 to 1997, 1,472 wives were killed by their husbands and 442 husbands were killed by their wives, a ratio of 3.3 to 1. It is noteworthy that there was a reported previous history of spousal violence in 56% of spousal homicide cases from 1991 to 1997. Alcohol consumption had been involved in 38% of these cases.

Sexual abuse of a partner typically involves forcing the victim to perform sexual acts against his/her will or physically attacking the sexual parts of the victim's body. It is difficult to assess the extent of sexual abuse in relationships for several reasons. First, offenders' relationship to the victim is not usually noted following charges of sexual offending. Second, the secretive nature of both sexual offending and spousal assault lowers the rate of reported sexual abuse in relationships. Statistics Canada (1999) included sexual assaults in their “Other Violent Offences” category, which totalled 5% of offences in 1997.

Emotional or psychological abuse is a broad category that involves controlling the victim through verbal means or creating an atmosphere of fear. In essence, psychological abuse incorporates all types of abusive behaviour that does not involve physical contact, and is seen as a component of all types of spousal violence. One type of psychological abuse that has achieved criminal status is criminal harassment (stalking). In 1997, 7% of all spousal abuse incidents reported to police were for stalking. Several other forms of abuse discussed in the literature are: economic abuse (encouraging and abusing economic dependence); isolation (controlling or limiting the victim's support network) or spiritual abuse (ridiculing or abusing the victim's spiritual or cultural beliefs). Although these are important factors to examine in spousal abuse situations, they can often be subsumed under the general group of emotional or psychological abuse.

RATIONALE FOR TREATMENT OF ABUSERS

Family violence has a damaging and costly impact on the community as well as a direct and sometimes tragic impact on the lives of affected spouses and children. The national survey on violence against women indicated that 45% of all women who had experienced violence had suffered an injury and, of these, 43% had required medical attention. Fifty-two percent had taken time off work as a result of injury (Statistics Canada, 1993). Statistics Canada estimates that health costs of injuries and chronic health problems caused by abuse amounts to a billion dollars every year (Day, 1994). Victims of domestic violence demonstrate increases in psychosomatic disease and drug and alcohol abuse. A significant portion of police time is allocated to intervening in family violence situations. Levens and Dutton (1980) coded taped call for police services in Vancouver and found that 13.5% were specifically for husband and wife disputes.

Family members are suspected in about one-quarter of all assaults against children. In the majority of cases of physical assault, fathers were the perpetrators (7 1%) and fathers were the primary perpetrators of sexual assault (97% of the cases). Family members were responsible for 76% of all child homicides. In 28% of these cases, there was known history of domestic violence. Children are also indirect victims of domestic violence as witnesses to the abuse. It is estimated that in as many as 80-90% of cases, the children know about the violence against their mother even if they do not directly witness the attack (Statistics Canada, 1999). In a recent analysis comparing children who have witnessed violence often, sometimes and seldom, to those who have never witnessed violence, researchers found that child witnesses were more likely to exhibit behaviours such as conduct disorder-physical aggression, emotional disorder, indirect aggression and property offences. In their recent review of the literature, Cunningham et al. (1998) noted that child witnesses often experience risk of injury as well as emotional trauma, reduced academic success and as adults their own families are often characterized by poor parent child communication. Suddermann and Jaffe (1999) noted that, behaviourally, children who witness violence become aggressive, non-compliant, irritable and easily angered. They also noted psychological problems of anxiety, depression, withdrawal, low self-esteem and an increase in somatic complaints. Socially, these children can have difficulty because of some of the above problems which often translates into academic problems. In their study of children in women's shelters, Suddermann and Jaffe found that 56% met the criteria for Post-Traumatic Stress Disorder, with most having some symptoms of this disorder. Finally, they also note the occurrence of “subtle” symptoms, such as inappropriate attitudes regarding conflict resolution and violence against women, condoning relationship violence, and hypersensitivity regarding problems at home and self-blame.

As adults, sons who are child witnesses of abuse are at an increased risk of becoming perpetrators of family violence and daughters are more likely to be victims. Straus, Gelles, and Steinmetz (1980) found that men who had witnessed wife assault in their families of origin had rates of battering three times greater than those who did not. Children in this study who were both witnesses of abuse and victims, were twice as likely to report an incident of spousal violence during the study year than those who did not (1 in 3). Clearly then, mounting an effective response to the problem of family violence would translate into reductions in individual and societal costs and would contribute to reductions in these costs in the next generation by breaking the intergenerational cycle of family violence.

THEORY DERIVED MODELS OF TREATMENT

A number of theoretical models have been proposed to explain the behaviour of abusive men. Sociological and sociobiological explanations provide broad analyses of culturally influenced or genetically programmed responses of abusive men but provide little direction on how to intervene clinically with individual abusers. Medical/ biological explanations cite evidence of links to organic brain damage that point to specific medical or clinical interventions for a limited number of perpetrators known to have such impairments.

Since the late 1970s the theoretical explanation that has had the most impact on the design of broadly based intervention programs for both male perpetrators and the victims of family violence is the feminist model. This approach points to the societal and political power imbalance between men and women as the key reason why men abuse women. The theory explains that the structure of patriarchal societies encourages the adoption of men's sense of entitlement to exert power and control over their families. This sense of entitlement justifies their use of a number of tactics such as the use of economic control, use of or threat of physical or sexual violence, and psychological tactics to maintain the power imbalance in their favour. In Canada, advocacy by women's groups for women and children who have been victims of male violence has contributed to an increase in public awareness of the problem, the development and funding of services for the victims of family violence as well as a greater sensitivity for the handling of these cases within the criminal justice system. Internationally, feminist analysis has contributed to the establishment of organizations that work to address the broad-based inequalities between men and women with many societies. Feminist based treatment programs for the victims of family violence have focussed on the empowerment of women and an analysis of the power dynamics. The Duluth treatment program is the most influential proponent of this model in the treatment of abusive men (Pence & Paymar, 1993). Their analysis of the tactics of men who batter, abusive men's attitudes toward their partners (the Power and Control Wheel) and the program's emphasis on learning egalitarian non-violent relationship strategies (the Equality Wheel) are now core components of most treatment programs for abusive men.

However, the research support for the theoretical basis of the approach and the treatment outcome literature on purely feminist based programs is thin. Sugarman and Frankle (1996), in a meta-analytic review of 29 studies of domestic violence, concluded that there was “limited support for the ideological component of the patriarchal theory of wife assault.” They found that, contrary to feminist theory, violent husbands were more likely to have an “undifferentiated” general schema, that is, they did not adhere to rigid sex role stereotypes and their attitudes toward women did not differ from non-violent husbands. Another criticism of the feminist explanation of family violence is that it cannot account for the high rates of violence among same sex relationships and evidence of women on men violence. What is more, treatment programs for male batterers that are strongly feminist and take an accusatory approach in working with participants may contribute to the high attrition rates reported in the field. Despite the important contribution the feminist based theory has made to the improvement of services for women and to a recognition of the contribution of political and economic disparity to domestic violence, it is not surprising that a single factor explanation cannot fully account for complex social phenomena nor that treatment programs derived from it cannot point to clear evidence of effectiveness.

A second theoretical orientation that has informed interventions with male batterers and with child witnesses is social learning theory. In this approach, domestic violence is conceived of as learned pattern of behaviour established through modelling patterns witnessed as children and in the society at large. The theory predicts that exposure to violent models in the home and in the popular culture increases the risk of a child becoming violent in his adult relationships. The treatment model inspired from this model downplays the role that individual psychopathology might play in the establishment of abusive patterns and instead directly targets the abusers' maladaptive responses to situations and events and retrains them to apply prosocial cognitive and behavioural responses. Evidence of the intergenerational trans-mission of violent patterns would appear to lend general empirical support for the theoretical model. However, it cannot account for the fact that most boys exposed to violent models will not become violent as adults. Kaufman and Zigler (as cited in Cunningham et al., 1998) noted decreased likelihood of adult abusiveness in child witnesses who had the love/support of one of his/her parents, supportive and loving adult relationships, acknowledged their experience of witnessing abuse, and committed to not being abusive in their own relationships.

Complete explanatory models should be multifaceted and include an analysis of how factors interact to contribute to domes-tic violence. Given that evidence points to a multi-modal treatment as being more effective in reducing criminal recidivism. It is reasonable to conclude that treatment programs to reduce relationship violence among criminal populations should also acknowledge the complexity of the origins of the problem by addressing multiple targets that are empirically shown to con-tribute to abusive behaviour.

A nested ecological model described by Dutton (1995) that explains domestic violence as multi-determined, influences the theoretical treatment approach adopted by the Correctional Service of Canada. The model, which is derived from the work of developmental psychologists and ethologists, provides a comprehensive explanation of intimacy violence. It considers the interactions between the broad social context, the perpetrator's intrapsychic features and the interpersonal context.

The model also points to appropriate targets to address in treating domestic violence perpetrators. It consists of four levels of social contexts, each influenced by the other.

1. Macrosystem. This first level consists of the broad attitudes and beliefs regarding wife assault that are held by one's culture. For example, the influence of patriarchy and the social and cultural prescriptions that endorse male aggression and men's power and control over women.
2. Exosystem. This level consists of social structures that influence the immediate context where the assault occurs. For example, work groups, friendships or other groups that connect the family to the larger culture. Work stress or the lack of social support could increase the risk for family violence. Association with other men who endorse violence toward women is another pathway that increases the risk that men will be abusive in their intimate relationships.
3. Microsystem. The third level consists of the family unit or the immediate environment within which the abuse takes place. This includes the level of conflict within the family unit, the factors that led up to, and the consequences of, the abuse; and
4. Ontogenetic Level. The last level is the individual component. Examples of individual factors related to family violence are: the perpetrator's developmental history, his possible experience of abuse and neglect as a child, his exposure to violent models, his degree of empathy, his ability to manage his emotions, his response to handling conflict and the level of anxiety over relationship changes. The individual response is influenced by his exposure to elements in the previous levels.

EFFECTIVE TREATMENT FOR ABUSIVE OFFENDERS

Effective intervention begins with a comprehensive theoretical model that points to treatment targets to address the multiple factors that influence relationship violence. For abusive offenders, we believe that the intervention should be consistent with broad features common to correctional programs that have the best outcomes in reducing recidivism. Meta-analytic and theoretical reviews have identified a cognitive-behavioural approach as the most effective treatment orientation in reducing criminal recidivism (Lipton, 1998; Lösel, 1995; Andrews & Bonta, 1994). There is limited research on the outcome of family violence treatment approaches with criminal populations, but one study found some support for a cognitive-behavioural approach over a process/psychodynamic approach with anti-social abusers (Saunders, 1996). In addition, meta-analytic studies have identified that correctional programs are more effective if they:

• are structured and focused, use multiple treatment components, focus on developing skills (social skills, academic and employment skills), and use behavioural and cognitive-behavioural methods (with reinforcements for clearly identified, overt behaviours as opposed to non-directive counselling focusing on insight, self-esteem, or disclosure);
• provide for substantial, meaningful contact between the treatment personnel and the participants (Sherman et al., 1997);
• ensure that treatment integrity is monitored to avoid program drift and ensure that service providers are adequately trained in the technique;
• select treatment targets that are dynamic factors related to risk (in this case attitudes and skills with due attention to substance abuse);
• apply interventions that employ active and participatory approaches such as role playing rather than passive didactic instruction;
• focus attention not only on highlighting the problem behaviour for the clients but also in assisting them to replace it with pro-social behaviour;
• bridge institution-based programs to community-based programs after release;
• arrange follow-up by using behavioural indicators of desired outcomes;
• use information learned from post-treatment follow-up to modify the program if needed.²

  (Reproduced, in part, from Cunningham et al, 1998)

The principal goal of a relationship violence program is the elimination of all forms of violent and abusive behaviour by offenders against their intimate partners. The programs seek to reduce the physical, sexual, emotional, psychological, and financial abuse of intimate partners. Although most family violence programs do not specifically provide instruction on parenting, or on prevention of abuse of the elderly, a secondary goal of most programs is the elimination of all violent and abusive behaviour in the family. Intermediate goals are to:

• develop perpetrators' insight into factors related to abuse;
• increase their awareness of the range of abusive attitudes and behaviours toward partners and children and the negative effects of these attitudes and behaviours in relationships;
• replace abusive attitudes and behaviours with non-abusive attitudes and behaviours; and
• develop a sense of responsibility for abusive and violent behaviours;

Some will argue that intervention should also improve the survivors' and children's well-being (Tolman & Edleson, 1995).

The principal targets for change in a cognitive-behavioural intervention for abusers are:

• The irrational or distorted attitudes and beliefs that influence the appraisal of the situation and/or allow the perpetrator to deny or reduce his responsibility for his violent or abusive behaviour. Cognitive-behavioural techniques are used to teach participants to analyse thinking patterns and then change the premises, assumptions and attitudes that underlie those thinking patterns (Edleson, 1996). Once the individuals are able to develop a critical awareness of beliefs underlying sexist and violent behaviour, they are then introduced to alternative beliefs, actions and behaviours. Examples of such distorted thinking are, images of masculinity that demand that men take control of family life, attitudes supportive of demeaning or abusing women, hostile attributions that construe neutral situations as ones called for aggression, unrealistic thinking that demands that all aspects of a relationships unfold in a specified way.
• The strong emotional responses that lead to aggression against partners. Many programs focus on anger or arousal of the perpetrator. Spouse specific anger/hostility, and not generalized anger has been found to predict relation-ship violence (Boyle & Vivian, 1996). Perpetrators learn how to monitor their arousal and recognize the cues signalling when they are losing their temper in their relationships. They then learn to apply anger/arousal control techniques such as relaxation, anger-down self-talk, reframing, etc. Throughout this process they learn to control their behaviour by controlling their arousal level. The same techniques can be used to monitor and manage other strong emotions related to relationship violence such as depression, jealousy and anxiety over relationship loss.
• Skill deficits. Effective programs will assist participants in addressing deficits in cognitive, coping and social skills required to deal with strong emotion and conflict or to forge healthier non-violent relationship patterns. Some cognitive skills that contribute to behavioural change are prosocial problem solving, learning how to anticipate positive and negative consequences of actions and restructuring problem thinking. High risk, or high need offenders may have social skills deficits that also need to be addressed. Learning effective communication and conflict resolution strategies reduces the risk of future assaults and increases the opportunity to build positive relationships.
• Problems in self-regulation. Teaching offenders how to anticipate difficult situations and inoculate themselves against negative experiences by applying relapse prevention techniques has become an innovation in correctional pro-grams since the mid 1980s (Pithers, 1990). Although there is not yet a body of outcome studies empirically supporting the use of a relapse prevention component in correctional treatment, it is entirely consistent with a systematic approach to risk management, particularly as a means of structuring the community follow-up component. This model helps the offender identify those factors that have contributed to his abusive behaviour pattern and points him to his internal resources (the appraisal component he has modified and coping skills he has learned) and his identified external resources (network of support) that he can rely on when confronted with stressful (high risk) situations. Follow-up in the community applying these techniques is particularly important when offenders have direct access to potential victims.

For a significant number of offenders, their abusive pat-terns will involve the abuse of substances, particularly alcohol abuse. An understanding of how substance abuse figures in the pattern of violence and the provision of specialized treatment for serious or chronic users is an essential component of effective correctional programs for abusers.

Appendix A provides an outline of the High Intensity Family Violence program recently implemented in institutions within CSC. The program is designed to treat high risk offenders with multiple needs; lower risk offenders may not require all elements of the program. We believe that it meets the criteria of effective programs for high-risk offenders and addresses the tar-gets relevant to this population. However, we do not yet have data evaluating the effectiveness of the program. The program has recently been accredited by an international experts' panel.

RESPONSIVITY ISSUES

Abusers, in general, are a heterogeneous group. They can vary significantly in education, social status, income and attitudes. Some abusers may only display violent behaviour in their intimate relationship(s). Others may have general deficits that result in their being violent in many settings. Some use alcohol as a disinhibitor; others do not. Some may be angry when they are violent; others may be calm, using violence as a instrument of control. Client characteristics contribute to differential outcomes following the initial assaultive incident. About one-third of men who assault their partners do not repeat the violence whether they receive treatment not (Rosenfeld, 1992). Others are multiple recidivists despite interventions. Treatment efforts should be responsive to client characteristics. The treatment approach appropriate for an apparently pro-social, educated and financially successful spousal abuser may be very different from a lower functioning pro-criminal abuser who uses violence in many contexts. Several researchers have developed typologies of abusers (Dutton, 1995, Holtzworth-Monroe & Stuart, 1994; Saunders, 1992; Hamberger, Lohr, Bonge, & Tolin, 1996). There are three broadly defined profiles they have identified: those who are generally violent and lacking in empathy (anti-social), those who are emotionally volatile and dependent (borderline personality organization) and those who experience discomfort dealing with intimacy but are not violent outside intimate relationships (over-controlled group; sometimes referred to as the non pathological group). Saunders' (1996) work provided some evidence that differential treatment approaches work better for some groups than others. The borderline and dependent abusers responded better to an unstructured supportive group while anti-social, manic and substance abusing men responded better to a more structured feminist cognitive-behavioural approach.

In correctional settings, the heterogeneity of the population is somewhat reduced. Among federal offenders, for instance, many offenders referred to treatment are not serving their sentences only for wife assault; they have general criminal histories as well. Preliminary research suggests there may be two broad profiles: offenders who are generally assaultive and criminally oriented and offenders who have specific histories of problematic relationships marked by attachment anxiety, jealousy and dependency. We have not yet reached the point in program development where we can determine the extent to which the treatment approach adopted within CSC will adequately address the needs of both profiles, but it is an area to consider in later phases of program development.

Another element of responsivity is the motivation of the abuser for self-change. Dropout rates for treatment in most community settings is high. Brown, O'Leary and Feldbau (1997) cite ranges of 40 to 60% for court mandated treatment programs and higher rates for self referred populations. It is obvious that not all perpetrators are equally ready to undertake personal change. While an effective treatment program helps the abuser recognize his responsibility for his violence, Murray and Baxter (1997) have discussed the counter-therapeutic effect of a confrontational and accusatory counselling style that relentlessly targets denial and minimization. The authors recommend applying the Trans-theoretical model (Prochaska & DiClemente, 1986), which proposes stage specific methods for intervening with clients. Abusers at the first stage, Precontemplators, would not yet see the necessity for changing their behaviour in relationships so attempting to engage them immediately in active treatment is not likely to be effective. Confrontational tactics may increase resistance. Motivational interviewing (Miller & Rollnick, 1991), on the other hand, is a collaborative style that engages the client by helping him to assess the costs and benefits of changing his behaviour. Other techniques recommended for Precontemplators are dramatic relief, in the form of testimonials from individuals who have completed treatment; films or books that present the effect of abusive behaviour and the value of change; consciousness raising information about the self and the problem; environmental reevaluation that helps the client assess how his problem affects others; and other techniques that encourage a belief in the individual's ability to change are appropriate methods to encourage Precontemplators to consider self-change (Levesque, 1998). It may be useful to provide a form of treatment primer consistent with the Transtheroretical model for offenders unwilling to participate in treatment.

A critical factor in treatment outcome is the quality of the relationship, or working alliance, forged between the client and the therapist or group facilitator. As mentioned above, the least effective are facilitators who are aggressive and authoritarian and use challenge and confrontation. Effective change agents are those who share with the client an understanding of the goals of treatment, share an understanding of the tasks required to get there and are able to forge a warm and supportive bond with the client (Bordin, 1994). The gender of the therapist may be especially critical in treatment of spousal abusers. Most group programs are co-facilitated by a man and a woman to take advantage of the opportunity it presents of modelling appropriate intergender relationships. The therapists can model a man-woman relationship based upon mutual respect. Furthermore, the woman therapist is able to speak more authoritatively on women's issues, and stands as a model that challenges “all women” statements (e.g., “Women are too emotional”). The male therapist is able to act as a model of prosocial male behaviours and challenge certain male stereotypes (e.g., “A man's house is his castle.”). Thus, by balancing the sex of the therapists, the team can better address issues that are raised in the group.

The forging of the working alliance with men of diverse ethnic backgrounds may be affected by the cultural competency of the therapist. Clinicians need to be sensitive to the cultural backgrounds of their clients. This sensitivity extends to under-standing the context of their behaviour and their personal beliefs that support violence and abuse within the family as well as the best values of the culture that support and promote positive images for families and intimate relationships.

ABORIGINAL OFFENDERS

The Correctional Service of Canada is committed to provide to Aboriginal offenders programs that are developed and delivered by Aboriginal experts. To the extent that Aboriginal offenders are not acculturated into main stream culture, specific programs should attend to differential learning styles and to the appropriateness of the selection of skills for the community to which the offenders will return. For example, Aboriginals from communities where there was relatively little contact with Euro-Canadian culture, who speak their native language and may have led more traditional life styles report feeling uncomfortable in group sessions that require disclosure of personal information and the expression of emotion. Eye contact in communication and the use of assertion skills are not adaptive skills for their social interaction (Waldram & Wong, 1994). Bicultural or assimilated aboriginal offenders may feel more comfortable in cognitive-behaviour group.

Aboriginal core programs integrate cultural and spiritual teachings such as the medicine wheel and ceremonies led by Elders, such as the use of sweetgrass, tobacco, and the sweat lodges with a cognitive behavioural treatment approach. These programs help Aboriginal offenders maintain or establish a link with their culture; to establish a route toward symbolic healing; and to identify key individuals who act as guides, healers and support on their release to the community.

In interventions to address family violence there are two major aspects unique to Aboriginal history that have had an impact of family violence and are likely to affect treatment: the impact of residential schools and post-colonial contact. The impact of residential schools on Aboriginal culture is well documented. For several decades Aboriginal children were removed from their home environments and placed in residential schools, punished for using their language and often prevented from extensive con-tact with their families. These children would have had much less exposure to effective family models (Taylor & Alksnis, 1995). They did not have the opportunity to observe their parents and grandparents deal with conflict, parenting or marital problems. In many cases, these children also suffered physical, sexual and emotional abuse from their caretakers or older residents. These would have been their models as they graduated and returned to their communities.

Treatment programs for Aboriginal offenders need to address the impact of residential schools. Offenders who may not have attended a residential school can often speak of its effect on their family and community. The inclusion of this material must be addressed in a very sensitive manner. Violence in relationships is always the responsibility of the perpetrators. While the negative cultural effects of residential schools cannot be used as a rationalisation for their abusive behaviour, it can be understood as a partial explanation.

Another issue that has an impact on family violence within Aboriginal communities is cultural abuse. This is a form of emotional psychological abuse where an individual's cultural/spiritual beliefs are denigrated. This type of abuse can enter the relation-ship if the partners are of different cultures or have different cultural beliefs. Although the partner may perpetrate this abuse, it is often reinforced by the larger culture. The use of residential schools to systematically acculturate Aboriginals is a prototypic example of how cultural abuse can occur at the societal level.

By examination of the cultural context, and the level of the offender's immersion in that culture, correctional staff may better serve the needs of their clients and assist in risk management in the community.

ASSESSMENT AND EVALUATION FRAMEWORK

Addressing spousal violence in correctional settings begins with the assessment of the perpetrator. There are three major purposes to assessment:

1. The prediction of risk for future spousal assault and the assessment of behaviour/attitudes that signal the return to abusive and violent behavioural patterns. Risk assessment identifies offenders who should receive more intensive service through informed supervision and treatment and follow-up.
2. The evaluation of treatment gain through the measurement of treatment targets such as attitudes and skill deficits associated with spousal violence.
3. The profiling of abusers through the assessment of characteristics. Profiling allows for post treatment evaluations to determine whether differential outcomes may be based on client variables.

Similar to other criminal offences (Andrews & Bonta, 1994), we typically rely on past behaviour to predict future behaviour in spousal abuse cases. The reliance on official records in cases involving intimate violence is problematic because of the very low reporting rate in such cases. However, unlike many other criminal offences, spousal abuse has an easily identifiable future victim. This makes monitoring easier when abusers are in the community.

The Spousal Assault Risk Assessment Guide (SARA) is one instrument that has been developed specifically to assess the risk for future spousal violence (Kropp, Hart, Webster, & Eaves, 1995). Items were selected based on retrospective analysis of histories of known spousal assaulters. The instrument is designed to be a case manager's guide to reviewing file and interview information. At this stage in the instrument's development, there are no cut-off scores to rank risk. However, after completing the guide the rater is required to assess the subject as at low, medium or high risk for future violence against his partner or others. The SARA has items associated with four major categories: criminal history, spousal assault history, characteristics of the current offence, and psychosocial adjustment. Generally speaking, the first three categories are relatively static, although there are some items, such as minimization or denial of spousal assault history that can change over time. Within the category of psychosocial assessment many of the items focus on recent functioning (e.g., recent relationship problems), while a few relate to historical issues (e.g., childhood witness/victim of family violence). The SARA allows for the identification of critical items. Through this process, the clinician can note items that he or she sees as central to their final risk rating. Although only a few items may have been endorsed on an individual's protocol, he may be rated as high risk because of one or two critical items. Conversely, an individual could have a high overall score but be rated lower risk because all the items pertain to incidents that occurred years ago.

Research on the SARA has confirmed that it performs better than the Psychopathy Checklist-Revised (PCL-R) in identifying future spousal assaulters. Moderate or high-risk ratings on the SARA point to the need for a comprehensive strategy to address the problem for that offender. In prison, individuals rated as high risk should be closely monitored during Private Family Visits. They should be referred to more intensive treatment options and closely supervised when released through regular home visits and contact with the spouse and children.

Other relevant measures include behavioural indices. A common rating scale used in spousal violence research is The Conflict Tactics Scale (CTS) and its revision (CTS2) (Straus, Hamby, Boney-McCoy, & Sugarman, 1996). The CTS was comprised of 19 items, subsumed under three scales: violence, verbal aggression and reasoning. The CTS2 included a revision of these scales: physical assault, psychological aggression and negotiation and the addition of two new scales: injury and sexual coercion. The CTS has been used for more than two decades to examine partners' use of physical violence, psychological violence and non-violent negotiation in dealing with relationship conflicts. The original intent of the CTS was for survey work in spousal violence (Straus, 1990).

Despite its widespread use, the CTS has come under a lot of criticism. Critics have noted that the instrument is limiting in that it focuses on conflict-related violence, has a limited set of violent acts, includes threats as violence and equates different violent acts (e.g., use of a weapon or a threat are both counted as one incident). Other researchers have criticized the choice of what is termed severe violence and minor violence. Straus (1990) responded to this criticism by noting that this division roughly parallels the distinction between common versus aggravated assault. Other criticisms focus on the scope of the CTS suggesting that the dynamics of family violence are ignored. Based on the results of the CTS, one cannot ascertain the context of the violence or who initiates violence. These issues become important in gaining a full understanding of the perpetrators.

Some of these criticisms seem to be addressed in the revised CTS2, others must simply be accepted as a limit of any assessment tool. Straus (1990) points out that researchers can add specific incidents that are important in their research. There has been an expansion of items in the CTS2, and it now examines sexual assault and injury (Straus, Hamby, Boney-McCoy, & Sugarman, 1996). Needless to say, this survey approach can provide researchers with much needed information. It presents conflict as a normal part of relationships that may encourage some respondents to be more forthcoming. It also allows for easy comparison between responses of each partner. Those using the CTS2 in clinical settings should be aware that it was not designed to examine relationship dynamics. Basically, it can be used as a relatively standard way of gathering data on incidents. It can also be used to examine changes in the frequency of incidents, a potential goal of therapy. Clinicians need also to examine the severity of the abuse through other means.

There are several self-report attitudinal measures used in the field. Typically these measures focus on assessing the extent to which men endorse sexist attitudes toward women or endorse the abusive and violent actions. Researchers recognize the short-comings of self-report measures because of their inherent susceptibility to impression management. To some extent, impression management can be statistically controlled though use of a measure of social desirability. However, even such an addition does not preclude the respondent denying or minimizing their attitudes or behaviours. Many researchers opt for combining official records, interview or rating methods with self-report and partner instruments to help increase the validity of their interpretations. A recent development in the assessment field is the use of standardized scenarios or vignettes to probe offenders' responses to hypothetical situations. Carefully constructed and scored, such measures can provide rich data and can mitigate the extent to which social desirability contributes to their responses.

Best practice in pre and post treatment assessment batteries would seek out convergent sources of data that tap behaviours and attitudes that are the target of treatment. Rating scales completed by facilitators should systematically assess the offenders' observed progress in treatment against the program goals (for example, an intervention specific Goal Attainment Scale (GAS-FV) (CSC, 1999).

Attitudinal measures (e.g., Shepard & Campbell (1992), particularly those that assess the participant's rationalizations around abusing women) combined with a social desirability measure (e.g., BIDR), assess changes in attitude towards women and women assault. Self-report measures based on the Transtheoretical model can assess readiness to change (e.g., the modified URICA, Levesque, 1998).

Objective tests assess the extent to which offenders have learned the knowledge content of the program. Participants' responses to scenarios or vignettes provide information on both skills development and attitude change and may not be as vulnerable to desirable responding as self-report measures. Profiling tools should be able to provide information on factors related to criminal recidivism and spousal abuse in particular: IQ, personality psychopathology, employment status as the time of offending, criminal history, age at time of offending and extent of problem with substance abuse.

Given that the primary goal of treatment is the cessation of violence against the partner, the principal treatment outcome variable of interest is a measure of repeated episodes of violence. Reliance on official records underestimates actual rates to such an extent that average base rates of recidivism are quite low. Most reports using official records have found that fewer than 20% reoffend after one to three years of follow-up. Such low base rates will require large differences between treated and untreated groups or very large sample sizes to produce significant treatment effects. When possible, follow-up interviews with the partner supply information on a range of abusive behaviours that official records lack. There are, however, several challenges in using this method as well. Obviously, outcome data is limited to those abusers who continue to co-habit. There are often difficulties tracking partners willing to participate in such follow-up interviews. Researchers have also noted that results of such surveys can be affected, among both the partners and male participants, by their increased awareness of the range of behaviours that are abusive after treatment.

TREATMENT EFFECTIVENESS

Since in most correctional settings, a group format is the most cost efficient, this section will only review the outcome on group programs. In community settings, however, couples counselling, family counselling and individual counselling are other formats for the delivery of interventions for relationship violence.

There are several problems that plague group treatment evaluation, and therefore, limit any firm statement about what works. Group interventions vary in their treatment approach and degree of structure. Some are primarily educational, some are unstructured self-help groups and more recently, court mandated pro-grams tend to combine a feminist analysis of power and control issues with a cognitive-behavioural approach. Non-representative samples are created through screening in volunteers and through group attrition. Many clinical studies have small samples, reducing statistical power. Studies use various definitions of abuse, different definitions of relationship, and various methods for reporting on outcome (official, self and partner reports). Most studies do not have a control group, although some use comparison groups.

Generally, evaluations have determined that most abusers (53-85%) stop their violence after treatment in follow-up periods ranging up to 54 months (Edleson & Syers, 1990; Dobash & Dobash, 1999). However, it is harder to make a definitive assessment of whether treatment provides incremental improvement beyond the deterrent effect of arrest.

Rosenfeld's (1992) review of mandatory treatment programs found that, on average, drop outs did just as well as those who attended treatment. He concluded that evidence to support the effectiveness of treatment was minimal. Dutton (1995), how-ever, has cited strong treatment effects for court mandated offenders. In a six-month follow-up, 16% of untreated abusers and 4% of treated offenders reoffended. Gains were maintained two and one-half years later when results indicated that 40% of untreated and 4% of treated men recidivated. The effects of treatment were also evident in samples of self reports from men and of women partners from the treated group which demonstrated that levels of violence and verbal aggression dropped after completion of the program. Conclusions of researchers based on their results have proved such contrary claims that they have been compared to competing political advertisements.

In other areas of correctional treatment, a major contribution to the confusing debate on specialized program effectiveness has been meta-analyses. Levesque (1998) examined the spousal assault literature using the meta-analytic method. However, reflective of the paucity of rigorous research, only 11 studies met the inclusion criteria. She found moderately significant improvement in the treatment group (Effect Size (ES) = 0.19; p<0.05) using official records. She found no differences (ES = 0.06, ns) between treatment and comparison groups when using partner reports. The overall results were heavily weighted by the results of one study (Harrell, 1991) which actually found that pro-gram participants did more poorly than an untreated comparison group. Since the study was well designed, it presents the alarming possibility that some interventions for male spousal assaulters may actually make them worse.

Since Levesque (1998) reviewed the outcome studies up to 1997, a number of important additional studies have added to the literature. Dobash and Dobash (1999) have recently evaluated non-equivalent criminal justice interventions with court mandated abusers. They found that all the criminal justice interventions resulted in reductions in violence but the treatment programs resulted in greater reduction and the reduction was sustained after one year.

In a 15-month follow-up study, Gondolf (1999) interviewed partners of men who attended abusive men's groups from four sites. The four programs were used because they were seen as well-established, maintained state standards, collaborated with women's programs, and use cognitive behavioural approach. One program was a 3-month pre-trial program, the second was a 3-month post-conviction referral program, the third was a 6-month post-conviction referral program and the fourth was a9-month post-conviction program that incorporated in-house substance abuse and individualized treatment. He found an over-all assault rate of 32%, a severe assault rate of 20% and a repeated assault rate of 19%. There were no significant differences among the four sites on overall assault rate; however, there were significant site differences in the other two categories. The 9-month program showed fewer severe assaults (12%) and fewer repeat assaults (11%). Since the general assault rates were similar across sites, based on a criterion of cost effectiveness, Gondolf endorses the 3-month intervention over the longer treatment. In the absence of a no treatment comparison group, however, the study cannot contribute to the question of whether treatment pro-grams in general are effective interventions.

Although reviews of outcome studies in the area can be con-fusing, one can generally conclude that success rates are only moderate at best but some men do seem to benefit from treatment (Cunningham et al., 1998). There is room for improvement in program design that could reasonably be expected to increase treatment effects. Many of the evaluated programs are short -- around ten sessions, only a few are more than 20 sessions. Based on his meta-analysis of correctional treatment, Lipsey's (1995) found that high intensity treatment, which he defined as those offering 100 hours of service, were more effective for high-risk offenders. None of the abuser programs we reviewed approached the recommended 100 hours of treatment. No formal descriptions are provided of methods that are used, if any, to engage poorly motivated clients and there is a paucity of programs designed and implemented for men from minority groups. Although few of the evaluations offered detailed descriptions of the program con-tent, it is not clear that any of them are providing relapse prevention models or are ensuring follow-up and maintenance for graduates of the program. In correctional settings we are well positioned to provide well informed supervision of graduates of the programs released to the community.

CONCLUSIONS

The above discussion is a brief introduction to a complex problem of spousal assault. There seems to be a small developing literature that is illuminating appropriate treatment in this area. Recently, programs apply an eclectic approach, linking power and control analysis with skills development under an over cognitive-behavioural rubric. Similarly to general correctional pro-grams (Andrews & Bonta, 1994), non-directive, unstructured and insight-oriented programs are not recommended. However, much more rigorous evaluation of programs is required before we will be able to point definitively to specific interventions that are more effective than others for this population. By linking the current research with what is known about general criminal offenders, we are able to develop plausible hypotheses regarding appropriate treatment and supervision for abusers in correctional settings.

---

---

REFERENCES

Andrews, D. A., & Bonta, J. (1994). The psychology of criminal conduct. Cincinnati, OH: Anderson.

Bordin, E. S. (1994). Theory and research on the therapeutic working alliance: New directions. In A. O. Horvath & L. S. Greenberg (Eds.). The Working Alliance (pp. 13-37). New York, NY: John Wiley.

Boyle, D. J., & Vivian, D. (1996). Generalized versus spouse-specific anger/hostility and men's violence against inmates. Violence and Victims, 11, 293-317.

Brown, P. D., O'Leary, K. D., & Feldbau, S. R. (1997). Dropout in a treatment program for self-referring wife abusing men. Journal of Family Violence, 12(4), 365-387.

Correctional Service of Canada (1999) Goal Attainment Scale-FV, Ottawa, ON.

Cunningham, A., Jaffe, P. G., Baker, L., Dick, T., Malla, S., Mazaheri, N., & Poisson, S. (1998). Theory-derived explanations of male violence against female partners: Literature update and related implications for treatment and evaluation. Paper presented for Reintegration Programs, Ottawa, ON: Correctional Service of Canada.

Day, T. (1994) Costs of violence against women. Preliminary results for Canadian Advisory Council on the status of Women. Ottawa, ON.

Dobash, R. E., & Dobash, R. P. (1999). Criminal Justice programmes for men who assault their partners. In C.R. Hollin (Ed.). Handbook of offender assessment and treatment. Chichester, UK: John Wiley and Sons.

Dobash, R. E., Dobash, R. P., Cavanagh, K., & Lewis, R. (1999) Changing violent men. Thousand Oaks, CA: Sage.

Dutton, D. (1995) The domestic abuse of women: Psychological and criminal justice perspectives (Rev. ed.). Vancouver, BC: UBC Press.

Dutton, D., & Hart, S. D. (1993). Risk markers for family violence in a federally incarcerated population. International Journal of Law and Psychiatry, 15, 101-112.

Edleson, J. L. (1996). Controversy and change in batterers' programs. In J. L. Edleson & Z. C. Eisikovitis (Eds). Future interventions with battered women and their families (pp. 154-169). Thousand Lakes, CA: Sage Publications.

Edleson, J. L., & Syers, M. (1990). Relative effectiveness of group treatments for men who batter. National Association of Social Workers, 26(2), 10-17.

Fitzgerald, R. (1999). Family violence in Canada: A statistical profile. Statistics Canada. Catalogue no. 85-224.

Gondolf, E. W. (1999). A comparison of four batterer intervention systems: Do court referral, program length, and service matter? Journal of Interpersonal Violence, 14(1), 41-61.

Hamberger, L. K., Lohr, J. M., Bonge, D., & Tolin, D. F. (1996). A large sample empirical typology of male spouse abusers and its relationship to dimensions of abuse. Violence and Victims, 11, 277-292.

Harrell, A. (1991). Evaluation of court ordered treatment for domestic violence offenders, final report. Washington, DC: The Urban Institute.

Holtzworth-Monroe, A., & Stuart, G. L. (1994). Typologies of male batterers: Three sub-types and differences among them. Psychological Bulletin, 116, 476-497.

Johnson, S. L., & Grant, B. (1998). Review of issues associated with serious spousal abuse among federally sentenced male offenders. Research Report R-84, Ottawa, ON: Correctional Service of Canada.

Kropp, P. R. (1998). Research related to the use of the SARA. Presentation at a Correctional Service of Canada working group. Ottawa, ON.

Kropp, P. R., Hart, S. D., Webster, C. D., & Eaves, D. (1995). Manual for the spousal assault risk assessment guide. (2^nd ed.). Vancouver, BC: The British Columbia Institute on Family Violence.

Levens, B. R., & Dutton, D. G. (1980). The social service role of the police: Domestic crisis intervention. Ottawa, ON: Solicitor General of Canada.

Levesque, D. A. (1998) Violence desistance among battering men: Existing interventions and the application of the transtheoretical model of change. Unpublished doctoral dissertation, University of Rhode Island, Rhode Island.

Lipsey, M. (1995). What do we learn from 400 research studies on the effectiveness of treatment with juvenile delinquents? In, What works: Reducing reoffending, edited by J. McGuire. Chichester, UK: John Wiley.

Lipton, D. (1998). How do cognitive skills training programs for offenders compare with other modalities: A meta-analytic perspective. Paper presented at the Stop and Think Conference, York, England.

Lockhart, L. L., White, B. W., Causby, V., & Isaac, A. (1994). Letting out the secret: Violence in lesbian relationships. Journal of Interpersonal Violence, 9(4), 469-492.

Lösel, F. (1995). The efficacy of correctional treatment. In What works: Reducing reoffending, edited by J. McGuire. Chichester, UK: John Wiley.

Miller, W. R., & Rollnick, S. (1991). Motivational interviewing: Preparing people to change addictive behavior. New York, NY: Guilford.

Murray, C. M., & Baxter, V. A. (1997). Motivating batterers to change in the treatment context. Journal of Interpersonal Violence, 12(4), 607-619.

Pence, E., & Paymar, M. (1993). Education groups for men who batter: The Duluth Model. New York, NY: Springer.

Pithers, W. (1990). Relapse prevention with sexual aggressors: A method for maintaining therapeutic gain and enhancing external supervision. In W. Marshall, D. Laws, & H. Barbaree (Eds.), Handbook of sexual assault: Issues, theories, and treatment of the offender. New York, NY: Plenum.

Prochaska, J. O., & DiClemente, C. C. (1986). Toward a comprehensive model of change. In W. E. Miller & N. Heather (Eds.), Treating addictive behaviors: Processes of change. New York, NY: Plenum Press.

Robinson, D., & Taylor, J. (1995). The incidence of family violence perpetrated by federal offenders: A file review study. Ottawa, ON: Correctional Service of Canada.

Rosenfeld, B. D. (1992). Court-ordered treatment of spouse abuse. Clinical Psychology Review, 12, 205-226.

Saunders, D. G. (1992). A typology of men who batter: Three types derived from cluster analysis. American Journal of Orthopsychiatry, 62(2), 264-275.

Saunders, D. G. (1996). Feminist-cognitive-behavioral and process-psychodynamic treatments for men who batter: Interaction of abuser traits and treatment models. Violence and Victims, 11(4), 393-414.

Shepard, M. F., & Campbell, J. A. (1992). The Abusive Behavior Inventory: A Measure of psychological and physical abuse. Journal of Interpersonal Violence, 7, 291-305.

Sherman, L. W., Gottfredson, D. C., MacKenzie, D. L., Eck, J., Reuter, P., & Bushway, S. D. (1997). Preventing crime: What works, what doesn't, what's promising. Washington, DC: Office of Justice Programs, U.S. Department of Justice.

Statistics Canada (1999). Family violence: A statistical profile. Ottawa, ON: Canadian Centre for Justice Statistics.

Statistics Canada (1993). The violence against women survey: Highlights. Ottawa, ON: The Daily, 18 November.

Straus, M. A. (1990). Ordinary violence, child abuse and wife beating: What do they have in common? In M. A. Straus & R. Gelles, (Eds.), Physical violence in american families: Risk factors and adaptations to violence in 8,145 families. New Brunswick NJ: Transaction Publishers.

Straus, M. A., Hamby, S. L., Boney-McCoy, S., & Sugarman, D. B. (1996). The revised conflict tactics scale (CTS2): Development and preliminary psychometric data. Journal of Family Issues, 17, 283-316.

Straus, M. A., Gelles, R. J., & Steinmetz, S. K. (1980). Behind closed doors: Violence in the American family. New York, NY: Doubleday.

Sudermann, M., & Jaffe, P. G. (1999). Children and adolescents who witness violence. New Directions in intervention and prevention. In R. D. Peters, R. McMahon, & D. A. Wolfe (eds.), Child abuse: New directions in prevention and treatment across a lifespan. (pp 5 5-78) Thousand Oaks, CA: Sage.

Sugarman, D. B., & Frankle, S. L. (1996). Patriarchal ideology and wife-assault: A meta-analytic review. Journal of Family Violence, 11, 13-40.

Taylor, J., & Alksnis, C. (1995). Models of family among aboriginal and non-aboriginal offenders. Ottawa, ON: Correctional Service of Canada.

Tolman, R., & Edleson, J. (1995). Intervention for men who batter: A review of the research. In S. R. Stith & M. A. Straus, (Eds.) Understanding partner violence: Prevalence, causes, consequences, and solutions. (pp. 262-273). Minneapolis: National Council on Family Relations.

Waldram, J. B., & Wong, S. (1994). Group therapy of Aboriginal offenders in a Canadian forensic psychiatric facility. American Indian and Alaska Native mental Health research, 6(2), 34-56.

Waldner-Haugrud, L. K., & Gratch, L. V. (1997). Sexual coercion in gay/lesbian relationships: descriptives and gender differences. Violence and Victims, 12(1), 87-98.

---

APPENDIX A

HIGH INTENSITY FAMILY VIOLENCE PROGRAM DESCRIPTION

Treatment targets

To increase offenders':

• Awareness of the consequences of their abusive behaviour
• Ability to respond non-abusively
• Ability to change abusive beliefs and behaviours
• Ability to identify high risk situations and to effectively manage these in the future

Treatment primer

Candidates will be assessed for readiness to change. For offenders who are identified as appropriate for the program, but are not ready to change (e.g., refusing treatment), a treatment primer will be used to prepare them for engaging in the treatment process. Long-term offenders who are not prioritized for treatment for several years may also be offered the treatment primer.

The treatment primer will consist of an information pack-age or resource kit designed to raise awareness of family violence issues and promote the value of addressing family violence concerns, in a non-confrontational manner. Resource materials will include fact sheets, books, videos, and testimonials and biographies of men who have changed. Offenders who use the resource materials will be given follow-up interviews with the program facilitators to discuss the materials.

Core program components:

Motivational enhancement
Goals:

• Increase interest in the program and motivation to change
• Develop group cohesion
• Develop trust in the facilitators and the therapeutic process
• Increase awareness of the extent and importance of the problem for each participant
• Develop personal goals

Psychoeducational component
Goals:

• Increase awareness and provide definition of abusive behaviours
• Develop understanding of the dynamics of family violence
• Increase understanding of both healthy and unhealthy relationship patterns
• Introduce relapse prevention and the ABC model, for incorporation into autobiographies
• Increase understanding of the link with substance use

Cultural component
Goals:

• Examine cultural influences on the development of beliefs and attitudes supportive of family violence
• Examine impact of transitions such as immigration, coping with racism on family dynamic
• Identify positive values in the culture of origin

Autobiographies
Goals:

• Develop understanding of early abusive relationship patterns in family of origin and their impact on current behaviour
• Develop understanding of personal abusive relationship patterns
• Identify personal risk factors and how they contribute to abusive behaviour
• Develop understanding of personal dynamics of abuse and identify personal abuse cycle
• Develop rationale for later presentation relapse prevention material

Skill building
Goals:

• Identify specific change targets, including thinking patterns, attitudes and beliefs, and behaviours that underlie abuse, using the ABC model
• Apply the ABC model to emotions management
• Develop skills to make targeted changes (e.g., challenge thinking errors, irrational beliefs, and controlling behaviour and replace with healthy prosocial alternatives)
• Develop social skills such as interpersonal problem solving, conflict resolution, and communication
• Practice skills using role plays and exercises
• Integrate skills into understanding of personal patterns
• Link skills to empathy building and maintenance of healthy relationships

Parenting
Goals:

• Identify the range of abusive behaviours that are child abuse
• Understand the impact of child abuse and being child witnesses of abuse on children
• Identify what abusers can do to assist child witnesses of abuse
• Discuss some aspects of non-abusive (nurturing) parenting
• Discuss how to manage high-risk situations that are triggered over co-parenting issues.

Relapse prevention and risk management
Goals:

• Identify personal risk factors and high risk situations for abusive behaviour
• Apply newly developed skills to coping with high risk situations using role plays and exercises
• Develop personal relapse prevention/risk management plans
• Share plans with partners
• Develop personal follow-up plans for the community, emphasising the importance of continued treatment, maintenance, and support services

Healthy relationships
Goals:

• Define healthy relationships
• Integrate all previous program material under the common theme of healthy relationships
• Apply program materials to the development of healthier relationships
• Review and closure

--------------------