Adapting DBT for Stalking Offenders

Adapting DBT for Stalking Offenders

This study demonstrated that stalking recidivism and stalking-related violence can be reduced through effective intervention. However, the specific type of intervention used may be less important than the rigor of the intervention. This is the bottom line of a recently published article in Law and Human Behavior. Below is a summary of the research and findings as well as a translation of this research into practice.

Featured Article | Law and Human Behavior | 2019, Vol. 43, No. 4, 319-328

Dialectical Behavior Therapy (DBT) for the Treatment of Stalking Offenders: A Randomized Controlled Study


Barry Rosenfeld, Department of Psychology, Fordham University;
Michele Galietta, Department of Psychology, John Jay College of Criminal Justice
Melodie Foellmi, Department of Psychology, Fordham University
Sarah Coupland, Department of Psychology, New York University
Zoe Turner, Department of Psychology, John Jay College of Criminal Justice
Stephanie Stern, Department of Psychology, John Jay College of Criminal Justice
Charity Wijetunga, Department of Psychology, Fordham University
Jacomina Gerbrandij, Department of Psychology, Fordham University
Andre Ivanoff, School of Social Work, Columbia University


The objective was to evaluate the relative efficacy of dialectical behavior therapy modified for stalking offenders (DBT-S) versus a cognitive– behavioral anger management intervention for the treatment of stalking offenders. We expected DBT to result in significantly lower rates of renewed stalking behavior and significantly greater improvements in impulsivity, aggression, anger, and empathy. We randomly assigned individuals charged with stalking-related offences (N = 109) to one of two study interventions: DBT-S and anger management. Recidivism (renewed stalking, violence, and other offences) was monitored for 1 year following the baseline assessment, and participants completed a battery of self-report questionnaires before and after treatment and at a 1-year follow-up assessment. We found relatively low rates of reoffence when compared to past studies of untreated stalking offenders in the U.S., but type of treatment had no impact on the likelihood of reoffence, nor did completion of the treatment program. Likewise, there was no between-groups difference in rates of treatment completion, or on changes in self-report measures. Intensive treatment focused on reducing problematic behaviors in stalking offenders may be effective regardless of treatment modality, but the mechanism by which treatment impacts criminal behavior is not yet clear.


stalking, offender treatment, dialectical behavior therapy, anger management, recidivism

Summary of the Research

“Few forms of abnormal behavior have aroused as much interest from criminal justice and mental health professionals over the past quarter century as stalking. Once considered a rare phenomenon limited to celebrities and public officials, the high frequency, and often-severe repercussions, of stalking behaviors are now well established. For example, several studies have demonstrated recidivism rates (typically defined as renewed stalking after an arrest or criminal charge) approaching or exceeding 50% for stalking offenders, and rates of violence between 18% and 40%. Despite the fact that most violence committed in the context of stalking does not result in serious physical harm, the impact of stalking on victims can be substantial, with high rates of psychological distress often resulting from long-term harassment. Indeed, anecdotal reports of long-term harassment (at times lasting more than a decade) have heightened interest in identifying strategies that might lead to desistence of pursuit behaviors. What remains unknown is whether any mental health interventions are useful in reducing or eliminating stalking behavior” (p. 319-320).

“To date, only one empirical study has systematically examined therapeutic interventions for stalking offenders. Rosenfeld, Galietta, and colleagues investigated the effectiveness of dialectical behavior therapy in a small sample of stalking offenders. The selection of DBT was based on research demonstrating a substantial rate of borderline personality disorder and cluster B personality disorders more generally among stalking offenders found in early descriptive studies of stalking offenders. Rosenfeld et al. (2007) contrasted participants who completed 6 months of DBT treatment with those who dropped out of treatment prematurely, as well as comparing the observed recidivism rate against published studies of stalking recidivism. Although these analyses supported the effectiveness of DBT in this sample, the uncontrolled nature of the study design limits the interpretation of significant treatment effects. Specifically, uncontrolled studies cannot evaluate the relative efficacy of an intervention compared to “treatment as usual.” The present study was intended to provide a more rigorous test of DBT for stalking offenders by comparing participants who received this intervention to a sample receiving a cognitive–behavioral anger management intervention that is analogous to those used in many criminal justice settings” (p. 320).

“Tomlinson (2018), following her review of the literature on DBT in correctional and forensic settings, concluded that ‘while the evidence-base for these programs is limited, modifications are generally consistent with best practices of offender rehabilitation according to the RNR model’ (p. 91). However, she noted that the ‘lack of rigorous, well-designed, randomized control trial studies in this area is problematic . . . [and] there is insufficient data to indicate that DBT is having an effect on recidivism that is greater than the effects of treatment as usual’ (p. 91). Thus, while a number of indicators point to the potential utility of DBT for offenders in general, and stalking offenders in particular, it is clear that further research addressing this issue is needed. Specifically, randomized clinical trials (RCTs) are needed in order to establish the efficacy of DBT, relative to other interventions in reducing stalking recidivism. The present study sought to fill this void, by randomly assigning stalking offenders to either a modified version of DBT adapted for stalking offenders (DBT-S) or a CBT-based anger management intervention, and monitoring recidivism (including stalking, violence, and criminal behavior more generally) and over a 1-year follow-up period” (p. 320-321).

“This study represents the first RCT of a modified version of DBT for the treatment of stalking offenders. Although a growing number of published studies have examined applications of DBT for criminal offenders, at times using experimental or quasi-experimental designs, none have used an RCT to examine the impact of DBT on recidivism or violent behavior in the community. Moreover, much of the research examining DBT in offender treatment settings, and indeed, much of the research on offender treatment more generally, has relied on small and/or convenience samples. In addition, this study utilized rigorous DBT training and adherence monitoring, both of which are uncommon even in DBT clinical trials. Finally, the examination of a range of outcomes beyond simply reducing risk of recidivism and violence, including a number of ‘personality’ variables such as empathy, impulsivity, and anger, represents yet another novel feature. In short, this study represents an important step forward in the examination of DBT in offender settings, utilizing a rigorous methodology and examining multiple outcomes in a sample that had adequate power to detect moderate treatment effects” (p. 325).

“Despite the rigor of this study methodology, the absence of any significant difference in our two treatment arms, DBT-S and a cognitive–behavioral anger management therapy, is clearly disappointing. Although the rates of reoffence suggest a very small, nonsignificant benefit from the more intensive DBT-S intervention, an exceptionally large sample would be needed to demonstrate a significant difference between treatment groups. More importantly, the magnitude of this effect size (ϕ = .10) does not reflect a clinically significant difference in treatment effects, further highlighting the apparent equivalency of the two interventions used in this study. There are a number of possible explanations for this outcome, including the most obvious: DBT-S is comparable, but not superior, to anger management for the treatment of stalking offenders. Although past research on stalking has highlighted a number of characteristics that would appear to support DBT as an ideal intervention for this population, stalking offenders appear to be far more heterogeneous than early research had suggested. Moreover, while DBT was developed as a treatment for women who engage in self-injury, anger management interventions have been more often used with men. However, prior studies have demonstrated roughly comparable effectiveness for DBT with men, suggesting that the impact of this intervention was not likely to have been hindered by the focus on a predominantly male offender sample” (p. 325).

Translating Research into Practice

“An equally plausible explanation for this study outcome, however, is that the comparison group used in this study was superior to the typical “treatment as usual” intervention. Although our original study design would have referred clients to a community-based, group anger management program, our need to ensure comparable practices with regard to attendance, treatment intensity, and termination policies forced us to create and operate our own comparison program. More importantly, we would not have been able to randomly assign referrals to a community-based treatment program—negating one of the most important features of any clinical trial. The result of this compromise was the utilization of an unusually high-quality anger management intervention that is atypical of TAU in virtually any community setting. Our anger management intervention was individualized (rather than group-based), and study therapists were highly motivated, extensively trained, and closely supervised in a structured CBT-based treatment model. Our program also included an enhanced risk management protocol, to ensure safety of study personnel, program clients, and potential third parties, further increasing the similarities between the treatment arms. Nevertheless, the lack of significant differences between these two interventions raises questions about whether the added cost of training and implementing DBT-S is worthwhile given the modest and nonsignificant improvements in the primary outcome variables” (p. 325).

“[T]he present study represents perhaps the first true RCT to examine DBT in a forensic setting, and the first rigorous test of any clinical intervention for stalking offenders. The seemingly modest recidivism rate observed suggests that clinical interventions may be beneficial for this population, but that the optimal treatment approach is not yet clear. For example, while enthusiasm regarding DBT has clearly grown within offender treatment settings, these findings suggest that using a structured CBT treatment approach with regular clinical supervision for providers may also be useful in the treatment of individuals who have engaged in stalking. Further research examining the relative effectiveness of these interventions, as well as the potential impact of ongoing risk assessment as an element of treatment, is clearly necessary” (p. 326).

Other Interesting Tidbits for Researchers and Clinicians

“Taken as a whole, the finding that both of our interventions resulted in comparably improved treatment outcomes has a number of important implications for offender treatment more generally. For example, the low rate of attrition highlights the potential for engaging an offender population that is not typically considered easy to engage or treat. Particularly given the nontrivial frequency of psychopathy and antisocial personality disorder diagnoses, the retention (completion) rate for a structured intervention such as DBT highlights the benefits of highly motivated, well-trained, and closely supervised treatment providers. In addition, the integration of ongoing risk assessment and management as a core element of treatment helped to sharpen the focus of treatment as well as to enhance safety of patients and staff. Nevertheless, given the evidence for a modest but significant association between treatment drop-out and psychopathy, greater attention needs to paid to retaining and treating this challenging subgroup of offenders” (p. 326).

Join the Discussion

As always, please join the discussion below if you have thoughts or comments to add!