Interpersonal style and the therapeutic alliance in sexual offender treatment

Offender rehabilitation has the potential to decrease criminal recidivism. However, many offenders do not benefit from rehabilitation efforts, with average reductions in recidivism rates approaching 20%; this gives scope for further room for improvement in rehabilitation efforts. Compared to risk assessment work and studies regarding treatment need, relatively little attention has been given to the process of treatment delivery, particularly as this may interplay between offender responsivity factors and the within-treatment change processes. The objective of the thesis was to examine the impact of sexual offender’s interpersonal style on therapeutic engagement and therapeutic gain. Three inter-related empirical studies were conducted, separately aiming to: 1) explore the impact of offender and therapist interpersonal style, including interpersonal complementarity, on the therapeutic alliance; 2) explore the frequency of ruptures and rupture repairs in the therapeutic alliance between offenders and treatment providers and examine the relationship between offender interpersonal style and the occurrence of ruptures and repairs; and to 3) explore the relationship between various offender characteristics and treatment process variables and treatment gain. To achieve these aims, self-report data were collected from 75 sexual offenders undertaking treatment over three time points, prior to the commencement of group treatment, soon after group treatment began, and prior to completion of the group treatment programme. Group facilitators also completed self-report measures soon after group treatment began. Treatment gain was assessed post-treatment, based on treatment completion reports and clinical notes. Correlational analyses, analyses of variance and multiple regression analyses were used to explore the relationship between interpersonal style, the therapeutic alliance, ruptures in the therapeutic relationship and treatment gain. Results showed that the interpersonal style of offenders and therapists impacted the treatment process via the therapeutic alliance. Offender interpersonal style and the therapeutic alliance were also associated with treatment gain. If either therapists or sexual offenders viewed the other as hostile or dominant, they were subsequently more likely to view the therapeutic alliance as weaker. The degree of complementarity between offenders’ and therapists’ interpersonal style, that is, whether their separate interpersonal styles matched (e.g., a dependent client was paired with a dominant therapist) did not improve the therapeutic alliance. Offender interpersonal style was also related to ruptures in the therapeutic relationship. Hostility and hostile-dominance in sexual offenders were related to the likelihood of a rupture occurring in the therapeutic relationship, however offender dominance was not. Offenders who did not experience a rupture in the therapeutic relationship had a stronger view of the therapeutic alliance than offenders who experienced a rupture in the therapeutic relationship that was not repaired. The strength of the therapeutic alliance was related to treatment gain (as measured through within-treatment change). However, the stage of alliance development and consideration of who was rating the alliance were important factors to take into account when predicting therapeutic outcome. Offender dominance and hostile-dominance were both related to treatment gain, whereas offender hostility alone was not. Therefore, it appears that dominant interpersonal traits impact offenders’ engagement in therapeutic change. In addition, sexual offenders with psychopathic personality traits were less likely to make treatment gain. These findings suggest that offender responsivity and treatment process are both important factors that should be taken into account when trying to understand treatment gain.