Dialectical Behavior Therapy (DBT), in partnership with the Risk Need Responsivity (RNR) principles, has the potential to reduce recidivism risk within the criminal justice system. Below is a summary of the research and findings as well as a translation of this research into practice.

Featured Article | International Journal of Forensic Mental Health | 2018, Vol. 17, No. 1, 72-95

A Theoretical and Empirical Review of Dialectical Behavior Therapy Within Forensic Psychiatric and Correctional Settings Worldwide


Monica F. Tomlinson, The University of Western Ontario


Cognitive-behavioral programs which are structured, skills-based, and risk-focused have been found to reduce recidivism rates by up to 55%. Dialectical behavior therapy (DBT) exemplifies all of these components, and has been rapidly adapted and implemented in correctional and forensic psychiatric facilities worldwide to reduce recidivism. Regrettably, the widespread implementation of
adapted DBT has outpaced the research on its effectiveness for this purpose. Thus, it is currently unclear whether these programs are meeting the rehabilitation needs of these systems. In the following article, a qualitative systematic literature review of all DBT programs within forensic psychiatric and correctional populations using the PRISMA statement guidelines is presented, along with a detailed exploration of how these programs align with best practices in offender rehabilitation, and whether they are effective in reducing recidivism risk. Results offer very preliminary evidence that DBT has the potential to reduce recidivism risk in criminal justice systems if applied within a Risk-Need-Responsivity framework.


Criminal justice policy, dialectical behavior therapy, DBT, offender rehabilitation, systematic literature review

Summary of the Research

“After decades of subscribing to a primarily punishment-based model, criminal justice systems around the world are increasingly calling for prison reform. Governments can no longer ignore the unequivocal evidence that punishment-based systems do not deter crime, are financially unsustainable, and fail to reduce recidivism. Unlike previous calls for prison reform, which lacked an empirically based alternative to punishment, the present movement is being supported by a wealth of research on effective offender rehabilitation. This research argues that criminological theories of risk and psychological approaches to mental health care can be synthesized and used to inform each other’s practice.” (p. 72)

“Among the third-wave cognitive-behavioral models, DBT has received particular attention in the forensic community. DBT is an intensive, structured, skills-based cognitive-behavioral program that enhances emotion regulation, distress tolerance, mindfulness, and interpersonal skills. DBT already includes several of the additional components of effective CBT programs (e.g., interpersonal skills, emotion regulation skills) and it has shown its effectiveness among difficult-to-treat populations, such as those with borderline personality disorder, eating, comorbid substance use disorders, psychopathy, and antisocial personality traits. Personality disorders and substance use disorders are specifically important to treat in forensic settings (compared to psychotic spectrum and mood disorders, for example) as they are significantly associated with recidivism (Crocker et al., 2015). Although this connection has not been made explicit in the literature, DBT also theoretically aligns with the most prominent and evidence-based risk reduction model in the recidivism literature, the Risk-Need-Responsivity (RNR) model. The RNR model posits that offenders should receive interventions that target the most acute risk factors for crime, that are matched in intensity to the person’s level of risk, and that are responsive to their individual needs.” (p. 73)

“The skills taught in DBT address many of the most significant risk factors for crime. For example, the module “interpersonal skills” can help individuals address many of the maladaptive interpersonal styles that lead individuals to developing procriminal associates. The “emotion regulation” module of DBT can assist individuals in better managing their emotions to help prevent aggressive or impulsive behavior. DBT is an intensive program (which would make it appropriate for high risk individuals), but can also be adapted to suit the risk level of offenders, as the four skills modules can be repeated and individuals can participate in as many “rounds” of DBT as needed. DBT is also derived from a cognitive-behavioral framework; which theoretically would enhance patient responsivity, according to Bonta and Andrews.” (p. 73)

“Arguably the most significant changes discussed in extant implementations of DBT within forensic settings pertained to the changes in skills training materials. Several sources extensively discussed how they adapted the material presented in the DBT skills training sessions to better serve the forensic setting. For practical reasons, many implementations changed the wording of the skills manuals to have simpler language (e.g., “Difficulties in Controlling Emotions” rather than “Emotional Dysregulation), less jargon (e.g., “Relationship Skills” instead of “Interpersonal Effectiveness”), more gender-neutral language (to account for the reality that forensic populations are predominately male), and more activities that apply to a custodial environment. The majority of programs suggested activities for reducing emotionality were not appropriate for a forensic con- text (e.g., having sex, arranging flowers, having a bubble bath), and thus had to be largely re-written. One source provided a completely re-written manual for a juvenile forensic population.” (p. 82)

“According to the “Risk” principle of the RNR model, program intensity should correspond to individuals’ risk levels. Since DBT is an intensive program, it would likely be most effective for high-risk populations. Of the 23 evaluated programs outlined in this review, six programs (26%) included only high- risk patients and two programs (9%) conducted separate DBT groups for participants at different risk levels. Seven programs (30%) did not report on the risk level of their participants, five (22%) programs included low or medium risk participants, and 3 programs (13%) included participants with varying risk levels.” (p. 87)

“The “Need” principle of the RNR model indicates that programs should target criminogenic needs. The adaptations and evaluations of DBT programs suggest that institutions are altering their programs to meet this principle. Of the “Big Four” criminogenic needs identified, the three dynamic needs (antisocial attitudes, antisocial personality traits, and procriminal associates) were all incorporated into DBT adaptations with forensic populations through changes in the DBT content. The majority of content changes to the standard DBT manual involved targeting crime-related thinking and behavior (which map onto the antisocial attitudes need). Additions to the manual included extensive crime-review sessions, groups for individuals with ASPD, and emotion regulation skills concentrated on increasing emotional responsivity and victim empathy, all of which map onto the antisocial personality traits need.” (p. 88)

“The “responsivity” principle of the RNR model asserts that rehabilitative programs should be tailored to the individuals’ learning styles, motivation, strengths, and abilities. Serin and Kennedy fleshed out the complexities of the responsivity component in great detail. They have emphasized a greater focus on understanding and addressing the “treatability” and motivation of offenders. An offender’s motivation for treatment can be assessed using self-report and staff-report measures of treatment motivation. An offender’s “treatability” can be determined by assessing: (1) the degree to which a program is appropriate to target the offender’s deficits; (2) the offender’s past experiences with the program (if any); (3) the programs/interventions that have been beneficial to the offender in the past; and (4) and whether there is reason to believe that the program may have contraindicated effects on the individual. Ideally, an offender would receive a program that is in line with his or her deficits, provided in a manner that has previously been beneficial for him or her, and has no expected contraindicating effects. Consistent with these recommendations, […] [I]ndividuals’ responsivity to their selected programming should be continuously monitored and programming should be adjusted as necessary. Creators of the RNR model, Bonta and Andrews, indicate that cognitive-behavioral treatments are most effective in maximizing the offender’s ability to learn from a rehabilitative intervention, and most beneficial to offenders who are “treatable” and motivated.” (p. 88)

“The state of DBT programs within forensic settings signals an encouraging, yet perhaps premature, move toward an empirically supported rehabilitation-focused approach to criminal justice policy. It is certainly hopeful to see forensic institutions implementing intensive psychotherapies with difficult-to-treat offenders. However, DBT has only recently been deemed an empirically-supported treatment for borderline personality disorder by the American Psychological Association (APA), and has not yet been deemed an empirically supported treatment for ASPD, or for forensic populations in general by Division 12 of the APA. Therefore, there are practical and ethical issues with the widespread implementation of DBT in forensic settings. The potential for iatrogenic effects in this highly vulnerable population certainly warrants caution and concern. To quell these concerns, forensic institutions around the world have been building an evidence base on which to support the use of DBT to reduce risk in forensic populations.” (p. 90)

Translating Research into Practice

“There is also some evidence that program fidelity was related to reduction in criminogenic needs. Across studies, the programs that adhered more faithfully to the program elements of DBT (e.g., skills training, individual counseling, consultation groups, and between-session coaching) were more effective in reducing criminogenic needs, such as poor impulse control, hostility, anger, emotional dysregulation compared to programs that did not implement all program elements of DBT. Furthermore, the programs that were applied faithfully within an RNR framework (targeted high risk individuals, criminogenic needs, and responsivity elements) were associated with reductions in institutional defiance and aggression more than programs that did not adhere to the RNR framework. It is possible, however, that programs which simply offered more services for longer were associated with better outcomes. More studies are needed to evaluate whether the length and intensity of programs are associated with better outcomes compared to shorter, less intensive programs. Overall, the findings from this review provide some indication that adapted implementations of DBT are able to reduce both risk and recidivism, and that programs which most successfully accomplish this goal adhered closely to standard DBT protocol (in terms of their program components) and fit within an RNR framework.” (p. 90)

“The present review of DBT programs within forensic settings demonstrates that forensic institutions are standardizing DBT within their institutions and evaluating their programs. Organizations such as Correctional Services Canada, the Colorado Mental Health Institute in Pueblo, and the Connecticut Department of Corrections, have written extensively on their standardized adaptations and preliminary evaluations. Many of these organizations have developed their own DBT manuals to better meet the needs of their populations. These manuals have integrated best practices for offender rehabilitation and incorporated the wealth of research on risk and rehabilitation.” (p. 90)

Other Interesting Tidbits for Researchers and Clinicians

“The present literature review of extant forensic DBT programs suggests 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. The general lack of rigorous, well-designed, randomized control trial studies in this area is problematic, however. Given that random assignment to groups was scarce, and several comparison groups were from different facilities or different time periods, there is insufficient data to indicate that DBT is having an effect on recidivism that is greater than the effects of treatment as usual. Further, due to important ethical limitations, most comparison groups are receiving some form of intervention, which precludes any evidence that DBT, alone, is reducing recidivism. Future research in this area is needed to improve the quality of studies, the size of research samples, and the fidelity with which programs are implemented. Such research can help to lead criminal justice policy into an era of prison reform that has the unprecedented luxury of standing upon empirically-supported approaches to offender rehabilitation.” (p. 91)

Authored by Kenny Gonzalez

Kenny Gonzalez is currently a master’s student in the Forensic Psychology program at John Jay College. His main research interest include forensic assessment, specifically violence risk. In the future, Kenny hopes to obtain a Phd in clinical forensic psychology and pursue a career in academia and practice.