Personality, Interpersonal, and Motivational Predictors of the Working Alliance in Group Cognitive Behavioral Therapy for Partner Violent Men

Transcription

1 Journal of Consulting and Clinical Psychology Copyright 2004 by the American Psychological Association 2004, Vol. 72, No. 2, X/04/$12.00 DOI: / X BRIEF REPORTS Personality, Interpersonal, and Motivational Predictors of the Working Alliance in Group Cognitive Behavioral Therapy for Partner Violent Men Casey T. Taft National Center for Posttraumatic Stress Disorder, VA Boston Healthcare System, and Boston University School of Medicine Christopher M. Murphy, Peter H. Musser, and Nancy A. Remington University of Maryland, Baltimore County Recent studies have demonstrated that the working alliance predicts treatment outcome for partner violent men. This study examined the influence of personality and interpersonal characteristics, motivational readiness to change, and demographic factors on working alliance formation among a sample of men (N 107) participating in a cognitive behavioral group treatment program for partner violence. Motivational readiness to change was the strongest predictor of the working alliance. Psychopathic personality characteristics also emerged as a strong (negative) predictor of the working alliance. Lower levels of borderline personality characteristics and interpersonal problems, self-referred status, married status, and higher age and income predicted higher working alliance ratings. The results support recent clinical efforts to address motivational readiness in programs for partner violent men. The working alliance, commonly conceptualized as the therapeutic bond between the client and therapist and their agreement on the goals and tasks of therapy (Bordin, 1979), has been positively associated with successful therapeutic outcome across a range of treatment populations and modalities (Martin, Garske, & Davis, 2000). Recent studies have obtained similar findings in group treatment for domestic abuse. In a study of couples versus gender-specific groups for partner violent men, Brown and O Leary (2000) found that husbands Session 1 working alliance ratings predicted reductions in physical and psychological aggression at posttreatment. More recently, we found the alliance to predict physical and psychological abuse levels during the 6 months after group cognitive behavioral treatment (CBT) in a predominantly court-mandated sample of men (Taft, Murphy, King, Musser, & DeDeyn, 2003). Therapist alliance ratings were particularly robust predictors of outcome. Casey T. Taft, National Center for Posttraumatic Stress Disorder, VA Boston Healthcare System, and Department of Psychiatry, Boston University School of Medicine; Christopher M. Murphy, Peter H. Musser, and Nancy A. Remington, Department of Psychology, University of Maryland, Baltimore County. This project was supported by an Individual National Research Service Award 1F31MH12234 to Casey T. Taft from the National Institute of Mental Health, by a grant from the Directed Research Initiative Fund of the University of Maryland, Baltimore County, and by National Institute of Mental Health Grant 1RO3MH56373 to Christopher M. Murphy. We gratefully acknowledge the 13 therapists who conducted groups during the period of this investigation and the generous support of the agency staff at the Domestic Violence Center of Howard County, Maryland. Portions of this work were presented at the Annual Convention of the Association for the Advancement of Behavior Therapy, Philadelphia, November Correspondence concerning this article should be addressed to Casey T. Taft, VA Boston Healthcare System (116B-2), 150 South Huntington Avenue, Boston, MA casey.taft@med.va.gov Given evidence that a positive working alliance is associated with reductions in abusive behavior, it is important to investigate the factors that contribute to alliance formation. To date, surprisingly little research has explored such factors (Connors et al., 2000), and no known studies have examined the development of a working alliance in treatment for partner violence. Therefore, we examined predictors of the working alliance among a sample of partner violent men in which the alliance outcome association had been previously demonstrated. Potential predictors were client variables across several domains, including personality disorder symptoms, interpersonal problems, motivational readiness to change, and demographic and referral factors. Approximately half of partner violent men in primarily courtmandated clinical samples possess prominent personality disorder traits, with even more showing relevant signs and features of these conditions (Gondolf, 1999; Hart, Dutton, & Newlove, 1993). Several recent studies have reported subtypes of partner violent men with prominent features of borderline and/or antisocial personality disorder (Holtzworth-Munroe, Meehan, Herron, Rehman, & Stuart, 2000; Waltz, Babcock, Jacobson, & Gottman, 2000). It is widely believed that individuals with borderline or antisocial features have difficulty forming a strong working alliance as a function of pervasive interpersonal difficulties (Frieswyk et al., 1986). The relationship between these personality features and the alliance has received scant empirical attention, however, and we are aware of no study that has examined interpersonal functioning variables as potential mediators of this association. Some psychotherapy studies have documented a direct link between both hostility and poor overall quality of current and past relationships with the working alliance (Kokotovic & Tracey, 1990; Moras & Strupp, 1982). Personality disorder characteristics and interpersonal problems may also influence the working alliance indirectly through motivational readiness to change, as individuals who possess rigid personality disorder traits and interpersonal styles may be less likely to see a need to change their problematic behavior and may, 349

2 350 BRIEF REPORTS therefore, be less amenable to positive alliance formation. Moreover, as others have argued (DiGuiseppe, Tafrate, & Eckhardt, 1994), developing a collaborative bond is inherently difficult with clients who persistently blame others for their interpersonal problems. Motivational readiness to change can be conceptualized within the transtheoretical model, which describes five stages of intentional behavior change. Each stage is characterized by distinct cognitive, attitudinal, and behavioral features (DiClemente et al., 1991; Prochaska & DiClemente, 1982). An individual s stage of change at any give time is thought to reflect his or her current motivational readiness for engaging in behavior change, a construct distinguishable from the alliance, which focuses specifically on the therapist client working relationship. Motivational readiness may be particularly relevant for partner violent individuals, as most are court-mandated to treatment and many deny a need for change on initial presentation (Daniels & Murphy, 1997). Some evidence indicates that court-referred clients are less motivated for change when compared with voluntary clients (Begun et al., 2003). A recent study of alcohol treatments found that client pretreatment motivational readiness was a strong predictor of the early alliance as rated by both clients and therapists (Connors et al., 2000). A final area of investigation involves demographic and background variables. These factors have been consistently associated with treatment attendance and dropout among samples of partner violent men (Daley & Pelowski, 2000), and it is plausible that such factors influence alliance formation. Ethnicity in particular may be relevant because of the experience of racism within the criminal justice system, a preference for informal support networks over formal therapy (Williams & Becker, 1994), and mistrust of majority group therapists (Taft, Murphy, Elliott, & Keaser, 2001). In partner violence treatment, minority ethnicity has been shown to predict dropout independent of factors such as education and income (Taft, Murphy, Elliott, & Keaser, 2001), and supportive communications regarding attendance have been shown to alleviate ethnic group differences in dropout (Taft, Murphy, Elliott, & Morrel, 2001). This study examined predictors of the working alliance among a sample of men in a CBT group for partner violence. It was predicted that demographic factors, psychopathic and borderline personality characteristics, court-related referral, total interpersonal problems, hostile dominant interpersonal problems, and motivational readiness to change would directly predict the alliance. It was also expected that interpersonal problems would mediate the association between the personality variables and the alliance, and the motivational readiness to change measure would mediate the relation between the nondemographic predictor variables and the alliance. Method Participants Participants were 107 men who sought counseling for perpetration of domestic abuse at a community-based agency. All had a documented problem with relationship abuse, as indicated by client or female partner reports of physical relationship aggression, extensive psychological abuse, or an arrest report with clear documentation of violence. The sample was 54% Caucasian, 38% African American, 3% Asian American, 2% Hispanic, 2% Native American, and 1% other. Thirteen (12%) of the participants were self-referred to treatment, and 94 reported either a pending legal case (8%) or a court-mandate (80%) to treatment. These 107 participants were drawn from a larger group of 163 potential participants who attended at least a portion of an intake session at the agency research site. Among potential participants, 14 refused research consent, 6 did not meet inclusion criteria, 2 met inclusion criteria but were deemed inappropriate for group treatment, 1 client s partner refused research assent, 13 dropped out of treatment during the intake process, and 20 were referred to group but failed to attend enough sessions to complete early working alliance ratings. There were no significant differences between participants and those excluded on any of the predictor variables of interest. All participants were assigned to a group CBT program (Murphy & Scott, 1996) that contained 16 weekly 2-hr sessions, and was conducted in a closed-group format by a male female therapist team. The majority of therapists were Caucasian (12 of 13). Group sizes ranged from 9 to 12 participants per group (M 10.8). A subsample (n 81) of those in the current study participated in a controlled clinical trial examining motivational interviewing as a pregroup preparation strategy (Musser, Murphy, & Taft, 2001). Among current study participants, 43 received two individual motivational interviews prior to group assignment, 38 received a structured control intake, and 26 received the normal clinical intake at the center, which had similarities to both of the other intake conditions. A more detailed description of the sample and treatment program is available elsewhere (Taft et al., 2003). Measures Psychopathic characteristics were assessed at intake using the Self- Report Psychopathy Scale II (SRP-II; Hare, 1990). The SRP-II is a 60- item self-report version of the Psychopathy Checklist Revised (PCL R; Hare, 1991). It was designed to assess Hare s two-factor model of psychopathy. Each item is rated on a 7-point scale, ranging from 1 (strongly disagree) to 7(strongly agree). Factor 1 measures underlying pathological interpersonal and affective personality traits. Factor 2 assesses antisocial features and criminal behaviors commonly associated with antisocial personality disorder. This study used a composite SRP-II scale intended to maximally correlate with PCL R total scores, as this scale has been found to possess the best psychometric properties. Hare (1991) revealed a correlation of.54 between the SRP-II and the PCL R among a criminal population (Hare, 1991), and significant (though unreported) correlations have been obtained in noncriminal settings (Forth, Brown, Hart, & Hare, 1996). In undergraduate samples, Lilienfeld and Andrews (1996) found their Psychopathic Personality Inventory to correlate highly (.91) with the SRP-II, and Zagon and Jackson (1994) found SRP-II scores to be higher among men, positively associated with narcissism, and negatively associated with empathy and anxiety. Borderline personality organization was assessed at intake with the Self-Report Instrument for Borderline Personality Organization (BPO; Oldham et al., 1985), a 30-item scale with subscales to assess identity diffusion, primitive defenses, and reality testing. An overall BPO composite was used for this study. The scale measures both clinical and subclinical levels of personality features associated with the diagnosis of borderline personality disorder (Oldham et al., 1985). The BPO scale was highly correlated with the Millon Clinical Multiaxial Inventory-II (Millon, 1987) Borderline Personality Scale (r.71), and significantly associated with partner-reported levels of emotional and physical abuse, as well as selfreported trauma symptoms, anger, and jealousy in a clinical sample of domestic abusers (Dutton & Starzomski, 1993), supporting its construct validity in this population. Interpersonal functioning was assessed at intake with the eight-subscale, 64-item circumplex scales of the Inventory of Interpersonal Problems (IIP; Alden, Wiggins, & Pincus, 1990; Horowitz, Rosenberg, Baer, Ureno, & Villasenor, 1988). IIP subscales include Domineering, Vindictive, Overly Cold, Socially Avoidant, Nonassertive, Exploitable, Overly Nurturant, and Intrusive. Alden et al. (1990) demonstrated that the eight subscales represent octants within a two-dimensional circumplex model. Participants report on a 5-point scale the degree to which each interpersonal problem applies to themselves. The subscales have been found to have adequate internal consistencies, with alphas ranging from.72 to.85 (Alden et al.,

3 1990). The IIP has been used as a structural validation for measures of interpersonal functioning by locating them in the circumplex space (Gurtman, 1992). A total composite score of interpersonal problems was used, as were composite measures of problems related to hostility dominance (sum of the Vindictive, Domineering, and Intrusive subscales), and friendliness submission (sum of the Overly Nurturant, Exploitable, and Nonassertive subscales). Motivational readiness to change was assessed immediately before the first group session with the Safe-at-Home Instrument (Begun et al., 2003) for assessing readiness to change intimate partner violence. This 35-item self-report measure was designed to assess the stages of change as described by the transtheoretical model, specifically as applied to individuals presenting for partner violence treatment. Confirmatory factor analyses in a large, multisite study of partner violent men in treatment indicated a clear three-factor structure: precontemplation (e.g., There s nothing wrong with the way I handle situations, but I get into trouble for it anyway ), contemplation (e.g., I want to do something about my problem with conflict ), and preparation action (e.g., I have a plan for what to do when I get upset ). Internal consistencies for eight-item versions of these subscales in the development sample were.59,.91, and.79, respectively. Concurrent validity was demonstrated through subscale correlations with referral status, self-efficacy for abstaining from partner aggression, and responsibility attributions for abusive behavior. A readiness to change composite was used, derived from composite scores of the eight items with the highest factor loadings for each of the three stage-of-change subscales, and computed as the sum of contemplation and preparation action minus precontemplation. We assessed the working alliance at Sessions 3, 5, 11, and 13 using the Working Alliance Inventory (; Horvath & Greenberg, 1989). The reliability and validity of the have been well-documented (Horvath & Greenberg, 1989; Tracey & Kokotovic, 1989). The long form of the (36 items) was administered to clients, who rated each cotherapist separately. The 12-item short form (Tracey & Kokotovic, 1989) was administered to group therapists. items are rated on a 7-point scale, ranging from 1 (never) to 7(always). The contains three subscales reflecting Bordin s (1979) bond (e.g., I feel comfortable with [name of person] ), task (e.g., We agree on what is important for me to work on ), and goal (e.g., I wish [name of person] and I could clarify the goals of our sessions ) components of the alliance. Global scores were analyzed because of past findings of high intercorrelations among the subscales (.77 to.96) and high internal consistency estimates for total scores (range.92.98; Taft et al., 2003), and because of previous validation studies demonstrating the superiority of a global factor (e.g., Tracey & Kokotovic, 1989). Ratings of and by the separate group therapists were averaged because of very high correlations between them for both client (.99) and therapist (.80) scores. ratings from Sessions 3 and 5 and Sessions 11 and 13 were averaged to reflect early and late ratings, respectively. Results Intercorrelations among the predictor variables are presented in Table 1. Consistent with mediational hypotheses involving personality disorder characteristics and interpersonal problems, both borderline and psychopathic characteristics were positively associated with hostile dominant interpersonal problems, and borderline characteristics were also associated with total interpersonal problem scores. Interestingly, friendly submissive interpersonal problems were negatively associated with psychopathic traits and positively associated with borderline traits. It was also expected that court-mandated status, personality disorder characteristics, and interpersonal problems would predict less motivational readiness, as motivational readiness was hypothesized to mediate associations between these factors and outcomes. Among these variables, only psychopathic characteristics were significantly associated with motivational readiness. Table 1 Intercorrelations Among Predictor Variables f BRIEF REPORTS 351 Variable No. of arrests 2. Age Education.33** Occupation.28**.24*.48** 5. Income.38**.42**.32**.45** 6. Employment a.04 b.12 b.09 b.30** b.29** b 7. Marital status c.19 b.25** b.13 b.15 b.45** b.04 d 8. Ethnicity e.12 b.08 b.21* b.03 b.08 b.10 d.07 d 9. Referral source f.36** b.03 b.16 b.23* b.19* b.09 d.14 d.06 d 10. SRP-II.25*.25* b.12 b.02 b.08 b 11. BPO b.17 b.01 b.30** b.23* 12. IIP: Total b.12 b.07 b.16 b.07.46** 13. IIP: hostile/dominant b.05 b.10 b.16 b.38**.41**.81** 14. IIP: friendly/submissive b.23* b.02 b.11 b.29**.37**.76**.32** 15. Motivational readiness b.10 b.04 b.04 b.29** Note. SRP-II Self-Report Psychopathy Scale II; BPO Self-Report Instrument for Borderline Personality Organization; IIP Inventory of Interpersonal Problems. a Unemployed 0; employed 1. b This is the point-biserial correlation coefficient. c Unmarried 0; married 1. d This is the phi coefficient. e Caucasian 0; minority group member 1. Self-referred 0; court referred or court case pending 1. * p.05. ** p.01.

4 352 BRIEF REPORTS Table 2 presents bivariate correlations between the predictors and ratings. Higher motivational readiness consistently predicted more positive ratings. Psychopathic characteristics were negatively predictive of early client ratings and both sets of late ratings. Total interpersonal problems, hostile dominant interpersonal problems, and referral source predicted late therapist ratings. Higher borderline characteristics were only significantly predictive of lower late client ratings. Married status and higher age were both predictive of positive client and therapist late ratings, and higher income was associated with higher late therapist ratings. Contrary to expectations, minority group status did not significantly predict ratings. We conducted two sets of mediational analyses in an attempt to explain the significant associations between psychopathy and the working alliance. First, analyses examined whether hostile dominant interpersonal problems mediated the association between psychopathic characteristics and late therapist. Consistent with the requirements for demonstrating mediation (Baron & Kenny, 1986), the correlations presented previously indicate that hostile dominant interpersonal problems were associated with psychopathic traits and late therapist. When entered together into a multiple regression equation predicting late therapist, both the effects of psychopathic characteristics, (93).15, pr.14, ns, and hostile dominant interpersonal problems, (93).16, pr.15, ns, were reduced to nonsignificant levels. A test of the significance of mediation (Kenny, Kashy, & Bolger, 1998) was nonsignificant (z 1.36, ns). The second set of mediational analyses tested whether motivational readiness mediated the significant associations between psychopathic traits and early client, late therapist, and late client. Consistent with the requirements for demonstrating mediation, correlations presented previously indicate that motivational readiness was associated with psychopathic characteristics and with these three sets of ratings. When entered together into a regression equation predicting early client, motivational readiness remained a significant predictor, (94).43, pr.43, p.001, the effect of psychopathic traits, (94).17, pr.19, ns, was reduced to a nonsignificant level, and the test of mediation was significant (z 2.33, p.05). Similarly, when predicting late therapist ratings, motivational readiness, (88).27, pr.26, p.05, and not psychopathic traits, (88).14, pr.14, ns, remained significant, and the test of mediation was marginally significant (z 1.81, p.07). Both motivational readiness, (88).38, pr.39, p.001, and psychopathic characteristics, (88).22, pr.24, p.05, were associated with late client alliance when entered together, and the test of mediation was significant (z 2.21, p.05). Discussion Two recent studies have demonstrated that a positive working alliance is associated with reductions in abusive behavior in treatment for partner violence (Brown & O Leary, 2000; Taft et al., 2003). The current project found a number of factors to be associated with positive working alliance formation among partner violent men, including high motivational readiness to change, low psychopathic personality characteristics, low borderline personality disorder traits, fewer total and hostile dominant interpersonal problems, self-referred status, married status, and higher age and income. Motivational readiness to change was particularly important for establishing a positive working alliance, as it was the only factor significantly associated with all four sets of working alliance ratings (i.e., both early and late ratings from both client and therapist perspectives). Counter to hypotheses, interpersonal problems did not mediate the association between personality disorder characteristics and the working alliance, nor did motivational Table 2 Associations Between Predictor Variables and Working Alliance Ratings Variable Early client Early therapist Late client Late therapist Primary predictor Variable SRP-II.31**.19.35**.21* BPO *.05 IIP: Total * IIP: hostile/dominant * IIP: friendly/submissive Motivational readiness.43**.41**.42**.33** Referral source a.03 b.08 b.07 b.21* b Demographic/background predictor No. of Arrests Age (years) *.22* Education Occupation Gross Yearly Income * Employment status c.01 b.08 b.02 b.05 b Marital status d.05 b.11 b.23* b.26** b Ethnicity e.11 b.06 b.11 b.05 b Note. Working Alliance Inventory; SRP-II Self-Report Psychopathy Scale II; BPO Self-Report Instrument for Borderline Personality Organization; IIP Inventory of Interpersonal Problems. a Self-referred 0; court referred or court case pending 1. b This is the point-biserial correlation coefficient. c Unemployed 0; employed 1. d Unmarried 0; married 1. e Caucasian 0; minority group member 1. * p.05. ** p.01.

5 BRIEF REPORTS 353 readiness to change mediate the association between referral source and the alliance. Motivational readiness to change, however, appeared to mediate the association between psychopathic characteristics and the working alliance. Contrary to expectations, client ethnicity did not predict alliance ratings. These findings are consistent with a recent study of alcoholism treatments (Connors et al., 2000) in which motivational readiness to change was associated with both client and therapist ratings of the working alliance and was the strongest predictor of the alliance among a large set of variables. Clients who enter treatment with low motivation to change are unlikely to agree with the therapist on the goals and tasks of treatment, and they may have trouble developing a warm and trusting therapeutic bond. Given that partner violent men are often considered to be treatment resistant and lacking in motivation for change (Pence & Paymar, 1993), the findings indicate that motivation is an important treatment target for this population. Recent investigations indicate that motivational enhancement techniques originally developed for substance-abusing clients (Miller & Rollnick, 2002) can be successfully adapted to enhance motivation and treatment involvement in partner violent men (Kistenmacher & Weiss, 2001; Musser et al., 2001; Taft, Murphy, Elliott, & Morrel, 2001). Court-mandated status was not significantly associated with motivational readiness to change, and these variables had independent associations with the working alliance. These findings appear to contradict previous research indicating that court-mandated clients have lower motivation to change than self-referred clients (Begun et al., 2003). Such inconsistent results most likely stem from the examination of overall readiness to change in the current study rather than the component scales for different stages of change. In the prior study, self-referred clients differed significantly from court-mandated clients in the contemplation of change, but not on other stage-of-change scales. In addition, among the current sample, the number of self-referred clients was relatively small, which may have attenuated associations with this variable. Potentially confounding the association between referral source and readiness to change was the fact that some of the participants in this study received motivational interviewing during the intake process, whereas others received only standardized, structured assessments. Inspection of the available data, however, indicates that the association between referral source and motivational readiness to change assessed prior to intake (r.10) was also very small. Moreover, associations between the other predictor variables and the working alliance were virtually unchanged when controlling for intake condition, suggesting that intake condition did not confound study results. Additional clinical implications arise from the link between personality disorder characteristics and the working alliance. Whereas borderline characteristics had a modest significant association with late client alliance ratings only, psychopathic characteristics were more strongly and consistently associated with the alliance. Previous research has demonstrated high levels of psychopathic and antisocial traits in this population (Murphy, Meyer, &O Leary, 1993). In other areas of investigation, individuals with these characteristics often have responded poorly to treatment (Rice, 1997). However, the results for partner violent men have been less convincing, with two recent studies showing no association between psychopathic characteristics and posttreatment violence recidivism (Gondolf & White, 2001; Remington & Murphy, 2001). Likewise, substance abusers with antisocial characteristics are often able to develop a working alliance, and their alliance ratings predict decreased drug use and higher employment at follow-up (Gerstley et al., 1989). Although it may be a difficult challenge, efforts to establish a strong working alliance with clients with antisocial or psychopathic traits may yield significant benefits. The current findings suggest that the alliance for these individuals may be facilitated by efforts to establish or enhance motivational readiness to change. The lack of an association between ethnicity and the working alliance was surprising, considering higher dropout rates found among minority group individuals and potential mistrust of majority group therapists (Taft, Murphy, Elliott, & Keaser, 2001). One possible explanation of these findings is that allianceenhancing interventions used at the agency research site, previously demonstrated to ameliorate ethnic disparities in attendance and dropout (Taft, Murphy, Elliott, & Morrel, 2001), may have similarly lessened differences in alliance formation in this study. It is also plausible that our approach to examining ethnic differences in the working alliance limited our ability to detect relationships. For example, perhaps our reliance on static working alliance ratings masked differences in alliance development over time. Another potentially fruitful area of future study involves client therapist ethnic match, which could not be adequately evaluated in the current sample because of limited therapist ethnic diversity. Some limitations of the study bear note. First, the modest sample size may have provided insufficient power to uncover the complex interplay of factors that contribute to alliance formation. In addition, a large number of associations were examined in this study, increasing the probability of spurious results. Another limitation involves response style effects emerging from the use of single-source data for some analyses. Interestingly, however, a greater number of predictor variables were associated with therapist alliance ratings than with client ratings, mitigating against explanations based solely on response tendencies, particularly given the previous finding that therapist alliance ratings were more predictive of abusive behavior outcome than were client alliance ratings (Taft et al., 2003). Finally, only client factors were examined as predictors of the alliance, leaving unexamined potentially important therapist factors such as empathy and directiveness, as well as factors that may emerge from the unique interaction of therapist, client, and treatment modality. References Alden, L. E., Wiggins, J. S., & Pincus, A. L. (1990). Construction of circumplex scales for the Inventory of Interpersonal Problems. Journal of Personality Assessment, 55, Baron, R. M., & Kenny, D. A. (1986). The moderator mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, Begun, A. L., Murphy, C. M., Bolt, D., Weinstein, B., Strodthoff, T., Short, L., & Shelley, G. (2003). Characteristics of the Safe at Home instrument for assessing readiness to change intimate partner violence. Research on Social Work Practice, 13, Bordin, E. S. (1979). The generalizability of the psychoanalytic concept of the working alliance. Psychotherapy: Theory, Research, and Practice, 16, Brown, P. D., & O Leary, K. D. (2000). Therapeutic alliance: Predicting continuance and success in group treatment for spouse abuse. Journal of Consulting and Clinical Psychology, 68,

6 354 BRIEF REPORTS Connors, G. J., DiClemente, C. C., Dermen, K. H., Kadden, R., Carroll, K. M., & Frone, M. R. (2000). Predicting the therapeutic alliance in alcoholism treatment. Journal of Studies on Alcohol, 61, Daley, J. E., & Pelowski, S. (2000). Predictors of dropout among men who batter: A review of studies with implications for research and practice. Violence and Victims, 15, Daniels, J. W., & Murphy, C. M. (1997). Stages and processes of change in batterers treatment. Cognitive and Behavioral Practice, 4, DiClemente, C. C., Prochaska, J. O., Fairhurst, S. K., Velicer, W. F., Velasquez, M. M., & Rossi, J. S. (1991). The process of smoking cessation: An analysis of precontemplation, contemplation, and preparation stages of change. Journal of Consulting and Clinical Psychology, 59, DiGuiseppe, R., Tafrate, R., & Eckhardt, C. (1994). Critical issues in the treatment of anger. Cognitive and Behavioral Practice, 1, Dutton, D. G., & Starzomski, A. J. (1993). Borderline personality in perpetrators of psychological and physical abuse. Violence and Victims, 8, Forth, A. E., Brown, S. L., Hart, S. D., & Hare, R. D. (1996). The assessment of psychopathy in male and female noncriminals: Reliability and validity. Personality and Individual Differences, 20, Frieswyk, S. H., Allen, J. G., Colson, D. B., Coyne, L., Gabbard, G. O., Horwitz, L., & Newsom, G. (1986). Therapeutic alliance: Its place as a process and outcome variable in dynamic psychotherapy research. Journal of Consulting and Clinical Psychology, 54, Gerstley, L., McLellan, T., Alterman, A. I., Woody, G. E., Luborsky, L., & Prout, M. (1989). Ability to form an alliance with the therapist: A possible marker of prognosis for patients with antisocial personality disorder. American Journal of Psychiatry, 146, Gondolf, E. W. (1999). MCMI-III results for batterer program participants in four cities: Less pathological than expected. Journal of Family Violence, 14, Gondolf, E. W., & White, R. J. (2001). Batterer program participants who repeatedly reassault: Psychopathic tendencies and other disorders. Journal of Interpersonal Violence, 16, Gurtman, M. B. (1992). Construct validity of interpersonal personality measures: The interpersonal circumplex as a nomological net. Journal of Personality and Social Psychology, 63, Hare, R. D. (1990). The Self-Report Psychopathy Scale-II. Unpublished Test, Department of Psychology, University of British Columbia, Vancouver, Canada. Hare, R. D. (1991). Manual for the Revised Psychopathy Checklist. Toronto, Ontario, Canada: Multi-Health Systems. Hart, S. D., Dutton, D. G., & Newlove, T. (1993). The prevalence of personality disorder among wife assaulters. Journal of Personality Disorders, 7, Holtzworth-Munroe, A., Meehan, J. C., Herron, K., Rehman, U., & Stuart, G. L. (2000). Testing the Holtzworth-Munroe and Stuart batterer typology. Journal of Consulting and Clinical Psychology, 68, Horowitz, L. M., Rosenberg, S. E., Baer, B. E., Ureno, G., & Villasenor, V. S. (1988). Inventory of Interpersonal Problems: Psychometric properties and clinical applications. Journal of Consulting and Clinical Psychology, 56, Horvath, A. O., & Greenberg, L. S. (1989). The development and validation of the Working Alliance Inventory. Journal of Counseling Psychology, 36, Kenny, D. A., Kashy, D. A., & Bolger, N. (1998). Data analysis in social psychology. In D. T. Gilbert, S. T. Fiske, & G. Lindzey (Eds.), Handbook of social psychology (4th ed., Vol. 1, pp ). Boston: McGraw-Hill. Kistenmacher, B., & Weiss, R. L. (2001, July). Motivational interviewing as a mechanism for change in men who batter. Presented at the International Family Violence Research Conference, Portsmouth, NH. Kokotovic, A. M., & Tracey, T. J. (1990). Working alliance in the early phase of counseling. Journal of Counseling Psychology, 37, Lilienfeld, S. O., & Andrews, B. P. (1996). Development and preliminary validation of a self-report measure of psychopathic personality traits in noncriminal populations. Journal of Personality Assessment, 66, Martin, D. J., Garske, J. P., & Davis, M. K. (2000). Relation of the therapeutic alliance with outcome and other variables: A meta-analytic review. Journal of Consulting and Clinical Psychology, 68, Miller, W. R., & Rollnick, S. (2002). Motivational interviewing: Preparing people for change (2nd ed.). New York: Guilford Press. Millon, T. (1987). Manual for the Millon Clinical Multiaxial Inventory (2nd ed., MCMI-II). Minneapolis, MN: National Computer Systems. Moras, K., & Strupp, H. H. (1982). Pretherapy interpersonal relations, patient s alliance, and outcome in brief therapy. Archives of General Psychiatry, 39, Murphy, C. M., Meyer, S., & O Leary, K. D. (1993). Family of origin violence and MCMI-II psychopathology among partner assaultive men. Violence and Victims, 8, Murphy, C. M., & Scott, E. (1996). Cognitive behavioral therapy for domestically assaultive individuals: A treatment manual. Unpublished manuscript, University of Maryland, Baltimore County. Musser, P. C., Murphy, C. M., & Taft, C. T. (2001, July). Motivating domestically abusive men for treatment. Presented at the International Family Violence Research Conference, Portsmouth, NH. Oldham, J., Clarkin, J., Appelbaum, A., Carr, A., Kernberg, P., Lotterman, A., & Haas, G. (1985). A self-report instrument for borderline personality organization. In T. H. McGlashan (Ed.), The borderline: Current empirical research (pp. 1 18). Washington, DC: American Psychiatric Press. Pence, E., & Paymar, M. (1993). Education groups for men who batter: The Duluth model. New York: Springer. Prochaska, J. O., & DiClemente, C. C. (1982). Transtheoretical therapy: Toward a more integrative model of change. Psychotherapy: Theory, Research and Practice, 20, Remington, N. A., & Murphy, C. M. (2001, November). Treatment outcomes of partner violence perpetrators with psychopathic and borderline personality characteristics. Presented at the annual meeting of the Association for the Advancement of Behavior Therapy, Philadelphia. Rice, M. E. (1997). Violent offender research and implications for the criminal justice system. American Psychologist, 52, Taft, C. T., Murphy, C. M., Elliott, J. D., & Keaser, M. L. (2001). Race and demographic factors in treatment attendance for domestically abusive men. Journal of Family Violence, 16, Taft, C. T., Murphy, C. M., Elliott, J. D., & Morrel, T. M. (2001). Attendance enhancing procedures in group counseling for domestic abusers. Journal of Counseling Psychology, 48, Taft, C. T., Murphy, C. M., King, D. W., Musser, P. H., & DeDeyn, J. M. (2003). Process and treatment adherence factors in group cognitive behavioral therapy for partner violent men. Journal of Consulting and Clinical Psychology, 71, Tracey, T. J., & Kokotovic, A. M. (1989). Interactional correlates of premature termination. Journal of Consulting and Clinical Psychology, 54, Waltz, J., Babcock, J. C., Jacobson, N. S., & Gottman, J. M. (2000). Testing a typology of batterers. Journal of Consulting and Clinical Psychology, 68, Williams, O. J., & Becker, R. L. (1994). Domestic partner abuse treatment programs and cultural competence: The results of a national survey. Violence and Victims, 9, Zagon, I. K., & Jackson, H. J. (1994). Construct validity of a psychopathy measure. Personality and Individual Differences, 17, Received October 28, 2002 Revision received March 24, 2003 Accepted July 2, 2003