Skip to main content
NCBI home page
Psychiatry, Psychology, and Law logoLink to Psychiatry, Psychology, and Law
. 2017 Aug 2;24(6):936–950. doi: 10.1080/13218719.2017.1347933

Responsive versus Treatment-Resistant Perpetrators in Batterer Intervention Programs: Personal Characteristics and Stages of Change

Pablo Carbajosa a,✉, Alba Catalá-Miñana a, Marisol Lila a, Enrique Gracia a, Santiago Boira b
PMCID: PMC6818265  PMID: 31984001

Abstract

This article aims to identify different personal characteristics in treatment-responsive and treatment-resistant perpetrators of intimate partner violence who completed a batterer intervention program (BIP). The sample consists of 105 perpetrators of intimate partner violence who were court-mandated to a community-based cognitive behavioral program. Perpetrators were classified by professionals as resistant or responsive to treatment based on the stage of change they reached upon completion of the program. The results show that before starting the intervention program, treatment-resistant perpetrators scored higher than treatment-responsive perpetrators in external responsibility attributions and attitudes toward violence in intimate relationships. No differences were found in personality disorders or psychological symptoms between the groups. However, longer program participation correlates with increasing differences between the two groups. The results suggest that targeting the personal characteristics which differentiate treatment-responsive perpetrators from treatment-resistant ones may help to increase the efficacy of BIPs.

Keywords: batterer intervention program, intimate partner violence, treatment-responsive perpetrators, treatment-resistant perpetrators, violent offenders

Introduction

Despite being a widely-used strategy for combating intimate partner violence (IPV), the efficacy of batterer intervention programs (BIPs) has been questioned. The meta-analyses that have evaluated these type of programs reveal limited efficacy (e.g., Arias, Arce, & Vilariño, 2013; Babcock, Green, & Robie, 2004; Eckhardt et al., 2013; Feder & Wilson, 2005). As a consequence, some new lines of study have highlighted the need to better adapt such programs to the variety of needs and characteristics that different perpetrators have (e.g., personality, level of risk, stage of change) in order to optimize coordination between the institutions and agents involved in the process, and to review the therapeutic approaches which the programs follow (Carbajosa & Boira, 2013; Gondolf, 2012; Lehmann & Simmons, 2009).

In this research context, one of the issues raising a lot of interest is the perpetrator's motivation to change, especially in court-ordered BIPs. Perpetrators’ motivational processes and their resistance to change have mainly been analyzed using the transtheoretical model (TTM) of behavioral change (Alexander & Morris, 2008; Eckhardt, Holtzworth-Munroe, Norlander, Sibley, & Cahill, 2008; Murphy & Ting, 2010). According to this model, people progress through five stages of change. At first, in the precontemplation stage, the individual has no intention to change and does not recognize the problem. In the contemplation stage that follows, the individual is aware that a problem exists and weighs pros and cons, but still does not commit to undertake any action to change. In the preparation stage, the individual is committed to change but is working out how to do it. Then in the action stage the individual makes active attempts to modify his or behavior and/or environment in order to overcome the problem. Finally, once the change has been made, it is consolidated in the maintenance stage (Prochaska & DiClemente, 1982; Prochaska, DiClemente, & Norcross, 1992).

This model has been applied in the area of IPV to classify individuals according to their stages of change, and in the capacity of the stages to predict the result of the treatments. The resulting research indicates that most perpetrators are in the precontemplation or contemplation stages at the start of a program (Eckhardt, Babcock, & Homack, 2004) – that is, they deny that they have a problem or they do not consider that there is a need to change (Alexander & Morris, 2008; Eckhardt et al., 2008), showing higher levels of resistance and a greater tendency to externalize responsibility for their violent behavior by blaming the system or the victim for their situation (Levesque, Velicer, Castle, & Greene, 2008; Lila, Oliver, Catalá-Miñana, Galiana, & Gracia, 2014). In addition, perpetrators in the precontemplation stage show fewer positive changes during the treatment with regard to empathy, communication skills and reducing their violent behavior (Scott & Wolfe, 2003). Similarly, unlike individuals in the later stages of the change process, those in the precontemplation stage are more likely to drop out of the program (Scott, 2004) and reoffend with both general crimes and aggression as reported by their victims (Alexander & Morris, 2008; Eckhardt et al., 2008).

In contrast, perpetrators who are in more advanced stages of change seem to establish a better therapeutic alliance and are more likely to embark on processes of change and show more pro-therapeutic behaviors during treatment, such as spontaneous assertions of responsibility for their abusive behaviors, encouraging other group therapy participants to take responsibility for their abusive actions, and claiming that group therapy is a worthwhile venture (Eckhardt et al., 2004; Semiatin, Murphy, & Elliott, 2013; Taft, Murphy, Musser, & Remington, 2004).

In light of the results of these studies, some authors differentiate between two groups of perpetrators: those with low motivation to change (in stages with low levels of problem recognition) and those with greater motivation (higher recognition and commitment to change; Alexander & Morris, 2008; Alexander, Morris, Tracy, & Frye, 2010; Crane & Eckhardt, 2013). The less motivated group of perpetrators holding negative attitudes have been labeled as resistant, and perpetrators with greater initial motivation to change as non-resistant (Scott, King, McGinn, & Hosseini, 2011). Perpetrators who are resistant to treatment or have low motivation to change are characterized by higher levels in attitudes of minimization, denial, and externalization of responsibility, and are associated with the stages of precontemplation and contemplation in which there is still no clear commitment to change (Boira, Del Castillo, Carbajosa, & Marcuello, 2013; Henning & Holdford, 2006; Levesque et al., 2008; Lila, Gracia, & Herrero, 2012; Murphy, Linehan, Reyner, Musser, & Taft, 2012). Characteristics such as personality traits are associated with resistant perpetrators in regard to their motivation to change. Typology studies along these lines have particularly associated borderline and antisocial traits with greater resistance to changing violent behavior (Eckhardt et al., 2008).

In sum, further research is needed into what characterizes the perpetrators who make progress in the stages of change throughout the intervention (responsive) and what distinguishes them from those who do not (resistant). Attitude factors such as initial motivation to change, level of acceptance of responsibility, the perception of the severity of the violence, and personality disorders and psychopathological symptoms appear to be key factors in differentiating between these perpetrators.

By classifying perpetrators completing a batterer intervention program into treatment-responsive and treatment-resistant groups, the first aim of the present study is to analyze whether or not at the beginning of the program they differ in terms of 1) demographic characteristics (age, education, income, marital status, and birthplace), 2) personality disorders and psychopathological symptoms, and 3) attributional and attitudinal variables (perceived severity of intimate partner violence and responsibility attributed to the legal system, the offender's personal context, and the victim). It was hypothesized that, at the beginning of the program, the group of resistant perpetrators would present less initial motivation to change, higher levels of external attribution of responsibility, lower perception of the severity of violence, and higher levels of pathology.

The study's second aim is to analyze the groups’ differences in their progress through the five stages of the TTM. It was expected that it would be possible to identify the points during the intervention program at which the group of responsive perpetrators – compared to the resistant perpetrators – tend to move out of the stages of low motivation to change (precontemplation and contemplation) and progress to the stage where they make a clear commitment to change (preparation) and subsequently begin carrying out actions to change their violent behavior (action or maintenance).

Method

Participants

The sample consists of 105 men who were court-mandated to participate in a community-based program for intimate partner violence offenders, which was facilitated by the University of Valencia (Lila, Oliver, Catalá-Miñana, & Conchell, 2014; Lila, Oliver, Galiana, & Gracia, 2013). The law on gender violence in Spain establishes that the prison sentence can be conditionally suspended if the batterers are sentenced to less than two years of prison and have no previous criminal record. As a requirement, they must complete a BIP. Initially, 147 consecutive cases were court-referred for treatment, although only 105 completed the intervention (28.60% dropped out during the treatment). Participants who did not attend three or more sessions were considered dropouts. Because this study excludes dropouts, the results are generalizable only to men who complete a full treatment program.

The average age of the participants who completed the program was 40.10 years (SD = 10.93), of which 35.90% were single, 26.20% were married or in a relationship, 24.70% were divorced, and 13.59% were separated. Regarding birthplace, the immigrant group accounted for 31.43%, of whom 12.38% are Latin American, 9.52% are European (except Spanish), 8.57% are African, and 0.95% are Asian. Regarding educational level, 6.70% had no education, 50.00% had elementary education, 33.70% had secondary education, and 9.60% had university education. Finally, their average incomes ranged from €6000–18000 per year.

Treatment

The Contexto Program was used for this study, a program which provides a community-based cognitive behavioral treatment for intimate partner violence offenders with activities based on the ecological model framework (Heise, 2011). The program consists of an assessment phase and a treatment phase. In the assessment phase, data from participants are collected through a set of questionnaires and in-depth interviews. The treatment phase is structured into six modules and runs over thirty two-hour group sessions (see Lila et al., 2013; Lila, Oliver, Catalá-Miñana, & Conchell, 2014). The groups are closed (no new members are enrolled after the program starts) and have 10 to 12 participants. Two doctoral-level psychologists lead each group. Briefly, in the first module (M1), the priority is to build a climate of trust within the group and establish its operating rules. In the second module (M2), basic concepts are explained regarding intimate partner violence, and some activities are run to target participants’ cognitive distortions and self-justifications for their situation (e.g., denial, minimization, victim-blaming), and help them to assume responsibility for their own behavior. From the third to the fifth modules (M3 to M5), the sessions aim to build up participants’ resources and skills, as well as to reduce risk factors at the individual and interpersonal levels. In the sixth module (M6), sessions target recidivism prevention and consolidate learning and training objectives (see Lila et al., 2013; Lila, García, & Lorenzo, 2010; Lila, Oliver, Catalá-Miñana, & Conchell, 2014).

Measures

Independent Variable

Treatment-resistant vs. treatment-responsive

The perpetrators were classified by two doctoral-level psychologists as resistant or responsive to treatment based on the stage of change after completing the treatment (M6). Treatment-resistant participants were considered to be those who showed they had not made a decision to take action after the treatment (and thus remained in either the precontemplation or contemplation stages of change). Treatment-responsive participants were considered to be those who showed they had made a decision to take action after treatment (and who thus had transitioned to the preparation, action, and maintenance stages of change). Therefore, after treatment, participants placed in the Precontemplation (n = 27) and Contemplation (n = 26) stages were labeled as treatment-resistant (n = 53), and participants placed in the Preparation (n =15), Action (n = 31) and Maintenance (n = 6) stages were labeled as treatment-responsive (n = 52). A dichotomous variable was thus obtained.

Dependent Variables

Stage of change assessment

In each case, the two independent doctoral-level psychologists specifically trained for this purpose rated each participant's stage of change (1 = Precontemplation, 2 = Contemplation, 3 = Preparation, 4 = Action, and 5 = Maintenance) based on the TTM (Prochaska & DiClemente, 1982; Prochaska et al., 1992). Questionnaires, individual interviews, and direct observation were used to provide judgments about the offenders’ stage of change (for a similar procedure, see Scott, 2004). The psychologists had a set of typical statements from each stage of change to use as a guide (see Appendix). To ensure assessment reliability, the two psychologists responsible for each treatment group assessed each participant independently. A high level of agreement was found between the psychologists (kappa = .70), according to the guidelines proposed by Landis and Koch (1977). The two psychologists then reached an agreement on the definitive stage in each case. In cases where agreement was not possible, an expert external observer was consulted. This process yielded a quantitative variable where 1 represents lower motivation to change and 5 represents higher motivation to change. This procedure was repeated throughout the intervention on seven occasions: before starting treatment (IN), and after completing each treatment module (M1, M2, M3, M4, M5, and M6).

The Millon Clinical Multiaxial Inventory-III (MCMI-III; Millon, 1994)

The MCMI-III is a self-report inventory consisting of 175 dichotomous items (true or false) to measure personality disorders. It comprises three modifying scales, eleven clinical personality pattern scales, three severe personality scales, seven clinical syndrome scales, and three severe syndrome scales (α = .65–.88). This inventory was completed in the assessment phase.

The Symptom-Checklist-90-Revised (SCL-90-R; Derogatis, 1992)

The SCL-90-R is a 90-item self-report tool used to assess current psychopathology on nine subscales and three global indexes. It is a five-point Likert-type scale where 0 = not at all and 4 = extremely (α = .81–.90). This inventory was completed in the assessment phase.

The Perceived Severity of Intimate Partner Violence Scale (PSIPVS; Gracia, García, & Lila, 2011)

The PSIPVS was evaluated using a scale in which participants had to rate the severity of eight hypothetical scenarios of intimate partner violence against women (e.g., ‘A couple has an argument, the man hits the woman, and later asks for forgiveness’; ‘A woman is threatened and insulted constantly by her partner, who sometimes pushes or hits her’) on a 11-point scale (0 = not severe at all to 10 = extremely severe). A general index is obtained by averaging the eight raw responses such that higher scores represent higher perceived severity of incidents of intimate partner violence against women (α = .87 for this study). This scale was completed in the assessment phase.

The Intimate Partner Violence Responsibility Attribution Scale (IPVRAS; Lila, Oliver, Catalá-Miñana, Galiana, & Gracia, 2014)

The IPVRAS is a 12-item scale designed to assess where the offender attributes the cause of his conviction for intimate partner violence against women. It is a 5-point Likert-type scale ranging from 1 = never to 5 = very often and comprises three dimensions: responsibility attributed to the legal system (e.g., ‘I am here because of an injustice’), responsibility attributed to the victim (e.g., ‘my partner's behavior and the way she treats me are the main reasons I am in this situation’), and responsibility attributed to the offender's personal context (e.g., ‘I am in this situation because of my jealousy’) (α = .53–.69 for this study). This scale was completed in the assessment phase.

Procedure

This study is part of a research project approved by the Ethics Commission on Experimental Research of the University of Valencia. Participants were recruited between 2011 and 2014. Participants completed a set of questionnaires in the assessment phase, including data used for the present study, as part of the protocol of Contexto Program. Later, individual in-depth interviews were conducted for further information, also as part of the protocol. These were 90-minute semi-structured interviews concerning several issues: biological and psychological conditions, initial attitude toward treatment, and resources for change (education, social support, responsibility attribution, etc.). After the interview, the initial stage of change was assessed (IN). Research consent was sought from the men during an orientation session. Participants were informed that the data obtained would have no impact on their legal situation and that confidentiality is guaranteed. The treatment was then applied. During the treatment phase, the participants’ stage of change was also rated after an individual interview at the end of each therapy module (M1–M6).

Data Analysis

Three sets of analyses were performed to detect differences in pre-treatment personal characteristics between groups, the first objective of the study. In all cases, membership of the treatment-responsive group or the treatment-resistant group was used as the independent variable. In the first set of analyses, independent analyses of variance (ANOVAs) were run to detect differences between groups in age, education, and income as dependent variables. Chi-square tests were also performed to detect differences in marital status and birthplace as dependent variables. In the second set of analyses, a multivariate analysis of variance (MANOVA) was conducted with pre-treatment scores in personality and psychopathological symptoms (the MCMI-III and SCL-90-R scores) as dependent variables. Due to the large number of dependent variables introduced in this case, the results were confirmed by discriminant analysis (Field, 2009) and by the distribution of the variables in MANOVAs, with fewer dependent variables based on Millon's (1994) classification. As the results did not differ, only results from the main MANOVA are shown (other data are available on request). Third, a MANOVA was conducted with pre-treatment scores in perceived severity of violence against women and responsibility attribution as dependent variables. To prevent Type I error when subsequent univariate contrasts were conducted, alpha levels were adjusted using Bonferroni correction (Harris, 1975).

To address the second objective, differences in the stage of change scores between the two groups over time were analyzed with a 2 (group) × 7 (time) repeated-measures MANOVA and subsequent univariate contrasts. Membership of the treatment-responsive or treatment-resistant group was used as the independent variable. The result of the stage of change assessments before treatment and after every module (IN, M1, M2, M3, M4, M5, and M6) were used as quantitative dependent variables (from 1 = Precontemplation to 5 = Maintenance). The Greenhouse–Geisser corrected solution was used in cases in which Mauchly's test indicated that the assumption of sphericity had been violated. To prevent Type I error, when subsequent univariate contrasts were conducted alpha levels were adjusted using Bonferroni correction (Harris, 1975).

Results

Pre-treatment Group Differences in Sociodemographic Characteristics

Table 1 shows pre-treatment differences between the treatment-responsive and treatment-resistant groups in sociodemographic characteristics. The differences between the two groups in terms of age, education, income, marital status, and birthplace are not significant.

Table 1.

Descriptive statistics and group differences in the pre-treatment sociodemographic variables.

  Treatment-resistant group (n = 53)
 Treatment-responsive group (n = 52)
    
  M DT M DT F η2
Age (years) 39.74 11.08 40.25 10.89 .06 < .01
Education levela 2.48 0.82 2.44 0.69 .07 < .01
Annual incomeb 4.12 2.15 3.81 1.76 .51 .01
Marital status %   %   χ2 φ
 Married/in a relationship 25.49   27.00   2.95 .17
 Single 31.37   40.38      
 Separated 11.76   15.38      
 Divorced 31.37   17.30      
Birthplace            
 Native 66.04   71.15   .32 .06
 Immigrant 33.97   28.85      
Open in a new tab

Note: a1 = no education, 2 = elementary education, 3 = secondary education, 4 = university education; b1 < €1800, 2 = €1800–3600, 3 = €3600–6000, 4 = €6000–12,000, 5 = €12,000–18,000, 6 = €18,000–24,000, 7 = €24,000–30,000, 8 = €30,000–36,000, 9 = €36,000–60,000, 10 = €60,000–90,000, 11 = €90,000–120,000, 12 > €120,000. Unequal variances between groups were considered according to Levene's test.

Pre-treatment Group Differences in Personality Disorders and Psychopathological Symptoms

A MANOVA was used to compare pre-treatment MCMI-III and SCL-90-R scores in the two groups. The multivariate effect is not significant, F (30, 71) = 0.83, p = .71, η2 = .26, thus no differences were found between the groups in any variable (Table 2). Discriminant analysis and the distribution of dependent variables in MANOVAs with fewer dependent variables confirmed no significant differences (data available on request).

Table 2.

Descriptive statistics and group differences in pre-treatment personality disorders and psychopathological symptoms.

  Treatment-resistant group (n = 53)
 Treatment-responsive group (n = 52)
    
  M SD M SD F η2
MCMI-III Schizoid 46.16 22.07 46.75 18.78 0.02 < .01
MCMI-III Avoidant 35.76 23.44 37.08 23.41 0.08 < .01
MCMI-III Depressive 33.92 23.90 36.71 26.38 0.31 < .01
MCMI-III Dependent 37.12 20.93 42.50 18.65 1.88 .02
MCMI-III Histrionic 50.66 18.30 48.65 19.12 0.29 < .01
MCMI-III Narcissistic 70.78 12.47 66.75 9.89 3.29 .03
MCMI-III Antisocial 47.76 22.26 46.19 20.94 0.02 < .01
MCMI-III Sadistic 41.64 25.15 39.08 21.84 0.30 < .01
MCMI-III Compulsive 61.42 19.36 61.90 19.40 0.02 < .01
MCMI-III Negativistic 40.04 22.19 40.44 23.37 0.01 < .01
MCMI-III Masochistic 33.04 23.55 34.21 23.60 0.06 < .01
MCMI-III Schizotypal 35.80 26.66 34.40 25.74 0.07 < .01
MCMI-III Borderline 33.78 23.73 36.50 25.90 0.31 < .01
MCMI-III Paranoid 51.10 25.42 46.04 25.84 0.99 .01
MCMI-III Anxiety 41.06 33.52 47.40 37.06 0.82 .01
MCMI-III Somatoform 34.04 28.10 28.92 28.01 0.85 .01
MCMI-III Bipolar: Manic 47.85 25.83 50.81 25.14 0.35 < .01
MCMI-III Dysthymia 29.26 25.10 29.67 26.71 0.01 < .01
MCMI-III Alcohol Dependence 46.12 27.03 48.87 23.26 0.30 < .01
MCMI-III Drug Dependence 45.78 27.40 47.81 26.78 0.14 < .01
MCMI-III Post-Traumatic Stress 29.50 26.41 36.46 28.59 1.62 .02
MCMI-III Thought Disorder 34.30 27.14 34.98 31.61 0.01 < .01
MCMI-III Major Depression 34.12 30.52 29.48 31.56 0.57 .01
MCMI-III Delusional Disorder 54.23 28.50 50.13 30.74 0.28 < .01
MCMI-III Disclosure 50.34 21.24 48.60 23.21 0.16 < .01
MCMI-III Desirability 79.54 15.28 78.08 14.73 0.24 < .01
MCMI-III Debasement 45.02 22.58 44.17 25.90 0.03 < .01
SCL-90 GSI 0.64 0.54 0.74 1.00 0.37 < .01
SCL-90 PSDI 0.46 0.68 0.44 0.69 0.10 < .01
SCL-90 PST 31.02 20.74 28.54 19.07 0.40 < .01
Open in a new tab

Note: GSI = Global Severity Index; MCMI-III = Millon Clinical Multiaxial Inventory-III; PSDI = Positive Symptom Distress Index; PST = Positive Symptom Total; SCL-90 = Symptom-Checklist-90-Revised.

Pre-treatment Group Differences in Perceived Severity of Intimate Partner Violence and Responsibility Attribution

A MANOVA was used to compare pre-treatment perceived severity of intimate partner violence and responsibility attributions. The multivariate effect is significant, F (4, 100) = 2.94, p < .05, η2 = .11, and subsequent univariate tests revealed significant differences in perceived severity of intimate partner violence, responsibility attributed to the legal system, and responsibility attributed to the victim. Specifically, the treatment-resistant group reported lower perceived severity of intimate partner violence, higher responsibility attributed to the legal system, and higher responsibility attributed to the victim. Pre-treatment group differences in responsibility attributed to the offender's personal context are not significant (Table 3).

Table 3.

Descriptive statistics and group differences in pre-treatment perceived severity of intimate partner violence and responsibility attribution.

  Treatment-resistant group (n = 53)
 Treatment-responsive group (n = 52)
    
  M SD M SD F η2
PSIPVS 68.62 13.00 73.56 7.77 5.55* .05
IPVRAS Legal System 13.91 4.34 11.98 3.91 5.69* .05
IPVRAS Personal Context 7.57 3.67 7.19 3.26 0.30 < .01
IPVRAS Victim 13.96 4.11 12.08 3.77 5.98* .06
Open in a new tab

Note: *p ≤ .05. IPVRAS = Intimate Partner Violence Responsibility Attribution Scale; PSIPVS = Perceived Severity of Intimate Partner Violence Scale.

Differential Progress in Stages of Change throughout Program: Treatment-Responsive Group vs. Treatment-Resistant Group

A 2 × 7 repeated measures MANOVA was used to examine the differences between the treatment-responsive and treatment-resistant groups, with groups as the between-subjects variable and time (IN, M1, M2, M3, M4, M5, and M6) as the within-subjects variable. A significant multivariate effect was found for time, F (4.35, 447.79) = 113.66, p < .001, η2 = .53, and for the interaction time × group, F (4.35, 447.79) = 53.64, p < .001; η2 = .34. Subsequent univariate tests revealed significant differences in the seven times between the two groups, and a gradual increase in effect size (Table 4). Independent repeated measures MANOVAs performed for each group show a significant main effect for time in the treatment-responsive group, F (4.29, 218.97) = 115.49, p < .001, and the treatment-resistant group, F (3.92, 204.03) = 9.88, p < .001. Specifically, the treatment-responsive group effect size, η2 = .69, is higher than the treatment-resistant group effect size, η2 = .16. Simple contrasts indicate that in the treatment-responsive group, stage of change scores increase significantly in all cases with respect to the pre-treatment assessment, from the IN–M1 contrast, F (1, 51) = 13.16, p < .05, η2 = .21, to the IN–M6 contrast, F (1, 51) = 816.75, p < .001. However, no differences were found in the IN–M1, IN–M2, and IN–M3 contrasts in the treatment-resistant group. Significant differences were found from the IN–M4 contrast, F (1, 52) = 14.96, p < .01, η2 = .22, to the IN–M6 contrast, F (1, 52)= 34.125, p < .001. Specifically, the IN–M6 contrast effect size is higher for the treatment-responsive group, η2 = .94, than the treatment-resistant group, η2 = .40.

Table 4.

Descriptive statistics and group differences in stages of change throughout the treatment.

  Treatment-resistant group (n = 53)
 Treatment-responsive group (n = 52)
    
  M SD M SD F η2
Interview 1.09 0.30 1.29 0.50 5.92* .05
Module 1 1.13 0.34 1.6 0.75 16.83*** .14
Module 2 1.13 0.34 1.87 0.82 36.22*** .26
Module 3 1.32 0.58 2.40 0.93 51.10*** .33
Module 4 1.38 0.49 2.77 0.90 97.56*** .49
Module 5 1.51 0.58 3.13 0.79 144.83*** .58
Module 6 1.49 0.51 3.83 0.62 451.46*** .81
Open in a new tab

Note: *p ≤ .05; *** p ≤ .001.

Figure 1 shows the differential progress in the stages of change throughout the treatment for both groups. The comparisons show that the treatment-responsive group has a higher stage of change average in all modules, and the differences increase as the treatment progressed. In the evaluation phase both the treatment-responsive and treatment-resistant groups were in the precontemplation or contemplation stages. Stage of change scores throughout treatment show that the treatment-responsive group had progressed to the contemplation stage by the end of module 3, reached the preparation stage by the end of module 5, and come close to reaching the action stage by the end of module 6. The treatment-resistant group remained in the precontemplation and contemplation stages throughout the treatment (see Figure 1 and Table 4).

Figure 1.

Figure 1.

Open in a new tab

Progress in the stages of change throughout the treatment for the treatment-responsive and treatment-resistant groups.

Discussion

The first aim of the study is to analyze whether the personal characteristics of the perpetrators in the treatment-responsive and treatment-resistant groups differed at the beginning of the intervention program (i.e., personality disorders, psychopathological symptoms, and demographic, attributional, and attitudinal variables). To meet this objective, perpetrators were classified by program therapists as resistant or responsive to treatment based on the stage of change they had reached on completing the intervention program (not reaching the ‘preparation’ stage vs. reaching the ‘preparation’ stage and beyond). Second, it was expected that it would be possible to observe the point at which the responsive perpetrators progressed out of the contemplation stage, as compared to the resistant perpetrators. The results generally confirm the first hypothesis on the initial differences between the resistant and responsive groups; specifically, the treatment-resistant perpetrators showed lower initial motivation to change than the treatment-responsive perpetrators. Before starting the intervention, the average group score on the stage of change for the treatment-responsive group was somewhat higher than that of the treatment-resistant group. The size of the effect of these differences is minimal, but in terms of recognizing their problem and their perspective on changing, none of the participants in either group had reached the preparation stage. All participants were either in the precontemplation or contemplation stages. These results are consistent with previous studies that place most perpetrators in the precontemplation and contemplation stages at the start of the program (Alexander & Morris, 2008; Boira & Tomás-Aragonés, 2011; Eckhardt et al., 2008).

No differences were observed between the two groups in terms of sociodemographic data. However, there do seem to be some differences – although not significant – in marital status. The largest difference is in the divorced category: the treatment-resistant group had a higher percentage of divorcees than the treatment-responsive group. The divorce process can be conflictive and cause animosity, especially if it is accompanied by allegations of domestic violence. As a consequence, going through this complicated process may give rise to greater resistance and defense mechanisms, although this question requires further exploration in future studies.

Further, contrary to the hypothesis, the data from the pre-treatment evaluation show that the treatment-resistant group did not differ from the treatment-responsive group in terms of personality disorders and psychopathological symptoms. These results are in line with those of Boira and Tomás-Aragonés (2011), where the perpetrators classified according to their stages of change at the beginning of the program also showed no significant differences in pathological symptomatology. One possible explanation is that in IPV aggressors the motivation to change may follow an independent process of personality. However, given the size of the sample and the number of personality variables incorporated in the analysis in this study, this result should be interpreted with caution.

Turning to the attributions and attitudes linked to violence, the treatment-resistant group attributed more responsibility for their violent behavior to the victim and the legal system in the pre-treatment phase. They also expressed more tolerant attitudes toward intimate partner violence (Gracia, Rodriguez, & Lila, 2015). It is possible that the members of the treatment-resistant group could tend to show greater resistance because their states of change are initially lower (Levesque et al., 2008). This result is especially relevant when considering that accepting responsibility for violent behavior is a priority aim of BIPs (Bowen, 2011; Lila, Oliver, Catalá-Miñana, Galiana, & Gracia, 2014; Scott & Strauss, 2007). No significant differences were found, however, between the groups with regard to the attribution of personal context to violent behavior.

Overall, the data from the initial evaluation seem to indicate that treatment-resistant perpetrators characteristically have lower levels of awareness of the problem, perceiving violence as less serious and maintaining external responsibility attribution, which may be variables associated with less progress in the stages of change throughout the treatment (Gracia et al., 2015; Levesque et al., 2008; Lila, Oliver, Catalá-Miñana, Galiana, & Gracia, 2014). Understanding these specific characteristics of the resistant group may provide valuable information for tailoring interventions. Perpetrators identified as resistant at the beginning of the program may require specific strategies designed to increase their initial motivation to change and reduce their level of responsibility externalization. Recent evidence suggests that resistant perpetrators with initial low motivation to change and high levels of externalization obtain better results when a motivational interview is included; this strategy entails a period of motivation-enhancing or specific treatment sessions which take place prior to the main treatment, during which individual differences in stages of change are taken into account (Alexander et al., 2010; Crane & Eckhardt, 2013; Kistenmacher & Weiss, 2008; Murphy et al., 2012; Musser, Semiatin, Taft, & Murphy, 2008; Scott et al., 2011).

Regarding the differences in their progress through the five stages of the TTM, the treatment-responsive and treatment-resistant groups were compared at seven points during the program. Once the treatment had started, the results revealed significant differences between the groups in their progress toward the next stage of change. The treatment-responsive group obtained higher average scores in the professionals’ evaluations of stages of change after each of the six treatment modules. These differences increased significantly until the end of the nine-month intervention period. Specifically, the treatment-responsive group – in contrast to the treatment-resistant group – reached the contemplation stage by the end of module 3 of the treatment. They then acquired a clear commitment to change (preparation stage) by the end of module 5 and finished the program with an average score close to the action stage. By contrast, despite having started the program in the same stages, the treatment-resistant group began and ended in the precontemplation or contemplation stage (stages in which the perpetrators had not begun any behavioral change initiatives). In general, these data seem to indicate that from the therapists’ point of view the progress of the treatment-responsive group into the preparation stage occurred halfway through the treatment. These results suggest that this could be the key moment to incorporate specific motivational strategies that ease the treatment-resistant group's passage into the preparation stage. One possible solution would be to incorporate a personalized motivational plan at the beginning of the program which is reevaluated throughout the treatment. Treatment-resistant perpetrators who remain in the precontemplation or contemplation stages at the middle of the program might benefit from individual motivational interview sessions that help them advance in their process of change.

This study has both strengths and limitations. Its limitations include classifying the perpetrators’ stage of change from the professionals’ point of view only. Although this measurement method was chosen as it is considered to be more accurate and avoids the high levels of social desirability characteristic of this population, including a self-administered questionnaire would have made it possible to observe possible discrepancies between these two types of measure. Nonetheless, previous studies have shown that the therapists’ evaluations are more accurate in predicting dropout from programs than self-reports by perpetrators (Scott, 2004). Continuing to advance on these types of measures and validating them in future studies could therefore be one avenue by which to increase the efficacy and predictive capacity of such tools in this population. Moreover, the validity of the stage of change assessment could be improved if raters were third-party observers rather than the therapists themselves. Another limitation to be considered is the small sample size related to the number of comparisons completed. The results should be interpreted with caution, as they could have limited power. Although some results have been confirmed in different ways, it would be advisable to verify them with larger samples in future studies.

Despite these limitations, one of the strengths of the study lies in its identification of characteristics that differentiate treatment-resistant perpetrators from those who advance in their stage of change process throughout the treatment. In addition, the progress of the stages of change was measured on seven different occasions, which made it possible to clearly identify the specific moments in the treatment at which transition from one stage to the next occurred.

In summary, improving the results of treatments involving men with more resistant attitudes to change continues to be a major challenge for both future research and interventions. The current ‘one size fits all’ programs have mainly addressed risk factors, and it seems necessary to continue advancing the design of programs that strengthen motivation to change and reduce resistance to assuming responsibility for violent behavior (Carbajosa, Boira, & Tomas-Aragonés, 2013; Stewart, Flight, & Slavin-Stewart, 2013). The results of the present study seem to suggest that for a group of perpetrators, factors such as the severity with which they perceive violence or the degree of external attribution of responsibility can influence the degree of progress toward the next stage of change during the treatment. The study also highlights the need to incorporate specific motivational strategies at different points in the program for change-resistant perpetrators. The current standard programs are useful tools for rehabilitating some perpetrators; however, further advances are needed in attending to more change-resistant perpetrators for whom these types of program seem to be less effective. The identification of treatment-resistant perpetrators by professionals may be crucial in designing differential motivational strategies for more effective treatment. It may be the case that in order to increase the general efficacy of programs, there is a need to develop strategies that are oriented to the individual characteristics of the most change-resistant aggressors.

Appendix

Appendix

Instructions: Please check off all descriptions that apply and then select the stage that you think best represents this man.

Stages of change Frequent statements
Precontemplation I am not violent
  She's the one that needs to change
  I've never had problems
  It's not true what she says
Contemplation Yes, I've been violent but she provoked me
  Yes, when I am angry I sometimes get violent. I know I've got a problem
  I'm more aware now that when I get angry, I hurt my partner
  My violent behavior has always got me into trouble
Preparation I think it would be good if they taught me how to control myself, but it's very difficult
  I've decided to come to treatment, I want to change and I need help
  I don't want to keep on hurting my family; from now on I'm going to control my temper, I can't go on like this
  This week I'm going to take note when I get angry and try to control myself
Action When I get nervous, I stop and think before I do anything even though it's not easy, but I'm still afraid of losing control
  I know I've got a problem with violence; this week I've been more careful and I've managed to control myself
  Now I talk about things and solve problems, not like before
  When I get uptight, I do what you told me
Maintenance I've learned how to resolve my conflicts and arguments in a different way; there are always solutions, the first thing is to relax, think about the consequences and look for alternatives
  If I see I'm going to lose it and I'm finding it difficult to handle I can always ask for help
  If I'm upset it's important to calm down and not drink any alcohol just then
  It's important to put myself in my partner's shoes and not only focus on the way I see things
  It's important to identify when I start to get angry and look for solutions before negative emotions build up
Open in a new tab

Funding Statement

This work was supported by the Spanish Ministerio de Economía y Competitividad [grant number PSI2011-25434].

Acknowledgements

This work was conducted in collaboration with the Home Office Prison Administration (Instituciones Penitenciarias. Ministerio del Interior).

References

  1. Alexander P. C., & Morris E. (2008). Stages of change in batterers and their response to treatment. Violence and Victims, 23(4), 476–492. [DOI] [PubMed] [Google Scholar]
  2. Alexander P. C., Morris E., Tracy A., & Frye A. (2010). Stages of change and the group treatment of batterers: A randomized clinical trial. Violence and Victims, 25(5), 571–587. [DOI] [PubMed] [Google Scholar]
  3. Arias E., Arce R., & Vilariño M. (2013). Batterer intervention programmes: A meta-analytic review of effectiveness. Psychosocial Intervention, 22, 153–160. [Google Scholar]
  4. Babcock J., Green C., & Robie C. (2004). Does batterers’ treatment work? A meta-analytic review of domestic violence treatment. Clinical Psychology Review, 23, 1023–1054. [DOI] [PubMed] [Google Scholar]
  5. Boira S., Del Castillo M., Carbajosa P., & Marcuello Ch. (2013). Context of treatment and therapeutic alliance: Critical factors in the intervention with court-ordered batterers. The Spanish Journal of Psychology, 16(40), 1–13. [DOI] [PubMed] [Google Scholar]
  6. Boira S., & Tomás-Aragonés L. (2011). Características psicológicas y motivación para el cambio en hombres condenados por violencia contra la pareja. International Journal of Psychological Research, 4(2), 48–56. [Google Scholar]
  7. Bowen E. (2011). The rehabilitation of partner-violent men. Chichester, UK: Wiley-Blackwell. [Google Scholar]
  8. Carbajosa P., & Boira S. (2013). Estado actual y retos futuros de los programas para hombres condenados por violencia de género en España. Psychosocial Intervention, 22(2), 145–152. [Google Scholar]
  9. Carbajosa P., Boira S., & Tomas-Aragonés L. (2013). Difficulties, skills and therapy strategies in interventions with court-ordered batterers in Spain. Aggression and Violent Behavior, 18(1), 118–124. [Google Scholar]
  10. Crane C. A., & Eckhardt C. I. (2013). Evaluation of a single-session brief motivational enhancement intervention for partner abusive men. Journal of Counseling psychology, 60(2), 180–187. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Derogatis L. R. (1992). The SCL-90-R. Baltimore, MD: Clinical Psychometric Research. [Google Scholar]
  12. Eckhardt C. I., Babcock J., & Homack S. (2004). Partner assaultive men and the stages and processes of change. Journal of Family Violence, 19(2), 81–93. [Google Scholar]
  13. Eckhardt C. I., Holtzworth-Munroe A., Norlander B., Sibley A., & Cahill M. (2008). Readiness to change, partner violence subtypes, and treatment outcomes among men in treatment for partner assault. Violence and Victims, 23(4), 446–476. [DOI] [PubMed] [Google Scholar]
  14. Eckhardt C. I., Murphy C. M., Whitaker D. J., Sprunger J., Dykstra R., & Woodard K. (2013). The effectiveness of intervention programs for perpetrators and victims of intimate partner violence. Partner Abuse, 4(2), 196–231. [Google Scholar]
  15. Feder L., & Wilson D. (2005). A meta-analytic review of court-mandated batterer intervention programs. Journal of Experimental Criminology, 1, 239–262. [Google Scholar]
  16. Field A. (2009). Discovering statistics using SPSS. London: SAGE publications. [Google Scholar]
  17. Gondolf E. (2012). The future of batterer programs: Reassessing evidence-based practice. Boston: Northeasern University Press. [Google Scholar]
  18. Gracia E., García F., & Lila M. (2011). Police attitudes toward policing partner violence against women: Do they correspond to different psychosocial profiles? Journal of Interpersonal Violence, 26(1), 189–207. [DOI] [PubMed] [Google Scholar]
  19. Gracia E., Rodriguez C. M., & Lila M. (2015). Preliminary evaluation of an analog procedure to assess acceptability of intimate partner violence against women: The partner violence acceptability movie task. Frontiers in Psychology, 6, 1567. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Harris R. J. (1975). A primer of multivariate statistics. New York: Academic Press. [Google Scholar]
  21. Heise L. L. (2011). What works to prevent partner violence? An evidence overview. London: STRIVE Research Consortium. [Google Scholar]
  22. Henning K., & Holdford R. (2006). Minimization, denial, and victim blaming by batterers: How much does the truth matter? Criminal Justice and Behavior, 33(1), 110–130. [Google Scholar]
  23. Kistenmacher B. R., & Weiss R. (2008). Motivational interviewing as a mechanism for change in men who batter: A randomized controlled trial. Violence and Victims, 23, 558–570. [DOI] [PubMed] [Google Scholar]
  24. Landis J. R., & Koch G. G. (1977). The measurement of observer agreement for categorical data. Biometrics, 33, 159–174. [PubMed] [Google Scholar]
  25. Lehmann P., & Simmons C. A. (2009). Strengths-based batterer intervention. A new paradigm in ending family violence. New York: Springer Publishing. [Google Scholar]
  26. Levesque D. A., Velicer W. F., Castle P. H., & Greene R. N. (2008). Resistance among domestic violence offenders. Violence Against Women, 14(2), 158–184. [DOI] [PubMed] [Google Scholar]
  27. Lila M., García A., & Lorenzo M. V. (2010). Manual de intervención con maltratadores. Valencia: PUV. [Google Scholar]
  28. Lila M., Gracia E., & Herrero J. (2012). Asunción de responsabilidad en hombres maltratadores: Influencia de la autoestima, la personalidad narcisista y la personalidad antisocial. Revista Latinoamericana de Psicología, 44(2), 99–108. [Google Scholar]
  29. Lila M., Oliver A., Catalá-Miñana A., & Conchell R. (2014). Recidivism risk reduction assessment in batterer intervention programs: A key indicator for program efficacy evaluation. Psychosocial Intervention, 23, 217–223. [Google Scholar]
  30. Lila M., Oliver A., Catalá-Miñana A., Galiana L., & Gracia E. (2014). The intimate partner violence responsibility attribution scale (IPVRAS). European Journal of Psychology Applied to Legal Context, 6(1), 29–36. [Google Scholar]
  31. Lila M., Oliver A., Galiana L., & Gracia E. (2013). Predicting success indicators of an intervention programme for convicted intimate-partner violence offenders: The Contexto Programme. The European Journal of Psychology Applied to Legal Context, 5(1), 73–95. [Google Scholar]
  32. Millon T. (1994). MCMI-III manual. Minneapolis, MN: National Computer Systems, Inc. [Google Scholar]
  33. Murphy C. M., Linehan E. L., Reyner J. C., Musser P. H., & Taft C. T. (2012). Moderators of response to motivational interviewing for partner-violent men. Journal of Family Violence, 27(7), 671–680. [Google Scholar]
  34. Murphy C. M., & Ting L. A. (2010). Interventions for perpetrators of intimate partner violence: A review of efficacy research and recent trends. Partner Abuse, 1(1), 26–44. [Google Scholar]
  35. Musser P. H., Semiatin J. N., Taft C. T., & Murphy C. M. (2008). Motivational interviewing as a pregroup intervention for partner-violent men. Violence and Victims, 23(5), 539–557. [DOI] [PubMed] [Google Scholar]
  36. Prochaska J. O., & DiClemente C. C. (1982). Transtheoretical therapy: Toward a more integrative model of change. Psychotherapy: Theory, Research and Practice, 19(3), 276–278. [Google Scholar]
  37. Prochaska J. O., DiClemente C. C., & Norcross J. C. (1992). In search of how people change: Applications to addictive behaviors. American Psychologist, 47(9), 1102–1114. [DOI] [PubMed] [Google Scholar]
  38. Semiatin J. N., Murphy C. M., & Elliott J. D. (2013). Observed behavior during group treatment for partner-violent men: Acceptance of responsibility and promotion of change. Psychology of Violence, 3(2), 126. [Google Scholar]
  39. Scott K. L. (2004). Stage of change as a predictor of attrition among men in a batterer treatment program. Journal of Family Violence, 19(1), 37–47. [Google Scholar]
  40. Scott K. L., King C., McGinn H., & Hosseini N. (2011). Effects of motivational enhancement on immediate outcomes of batterer intervention. Journal of Family Violence, 26(2), 139–149. [Google Scholar]
  41. Scott K. L., & Straus M. (2007). Denial, minimization, partner blaming, and intimate aggression in dating partners. Journal of Interpersonal Violence, 22, 851–871. [DOI] [PubMed] [Google Scholar]
  42. Scott K. L., & Wolfe D. A. (2003). Readiness to change as a predictor of outcome in batterer treatment. Journal of Consulting and Clinical Psychology, 71(5), 879–889. [DOI] [PubMed] [Google Scholar]
  43. Stewart L. A., Flight J., & Slavin-Stewart C. (2013). Applying effective corrections principles (RNR) to partner abuse interventions. Partner Abuse, 4(4), 494–534. [Google Scholar]
  44. Taft C. T., Murphy C. M., Musser P. H., & Remington N. A. (2004). Personality, interpersonal, and motivational predictors of the working alliance in group cognitive-behavioral therapy for partner violent men. Journal of Consulting and Clinical Psychology, 72(2), 349–354. [DOI] [PubMed] [Google Scholar]

Articles from Psychiatry, Psychology, and Law are provided here courtesy of Taylor & Francis

RESOURCES