Correctional Service Canada
Symbol of the Government of Canada

Common menu bar links

FORUM on Corrections Research

Warning This Web page has been archived on the Web.

Programming for violent offenders

by Ralph C. Serin and Denise L. Preston1
Research Branch, Correctional Service of Canada

The preoccupation with violent offenders has heightened following the emphasis on risk appraisal over the past decade. It is therefore not surprising that in addition to changes in sentencing and policy, correctional jurisdictions are now attending to the treatment and management of high-risk and violent offenders. This article focuses on interventions and programs for violent offenders that are intended to reduce recidivism. Further, these programs can be linked to the specific treatment needs of violent offenders rather than criminality more generally.

Defining violent offenders

A major impediment in the treatment of violent offenders has been confusion over their definition. Violent offenders are usually defined in terms that are not mutually exclusive such as criminal convictions (e.g., assaults), attitudes (e.g., hostility), emotions (e.g., anger), and victim selection (e.g., spousal assault). The failure to specifically delineate types of violent offenders has obscured the identification of treatment needs and confounds program effectiveness research.2 For example, predominantly instrumentally aggressive clients are unlikely to show substantive gains in an arousal management-based anger control program. Further, even observable and measurable changes by an offender (e.g., knowledge of anger principles), may well be unrelated to reductions in future violence simply because the domain was not criminogenic for that particular offender. Upon evaluation, such a program might then be considered to be ineffective. More accurately, it should be considered ineffective for certain types of violent offenders.

Treatment needs and targets

Literature reviews of risk factors in chronically violent or aggressive individuals yield such problems as:

  • hostility
  • impulsivity
  • substance abuse
  • major mental disorders acute symptoms
  • antisocial personality, psychopathy
  • social information-processing deficits
  • experience of poor parenting
  • neglect as a child.

These factors or treatment targets can be organized into domains and compared among different types of violent offenders to demonstrate the need for matching offenders’ treatment needs with program content. These five domains can be related to the expression or inhibition of violent behaviour. These domains are:

  • competence (social skills and empathy)
  • arousal (anger)
  • schema (aggressive beliefs and hostile attributions)
  • self-regulation (impulsivity)
  • anxiety (neuroticism).

Tolan and Guerra have described four different types of violent offenders, which include predatory, relationship, situational, and psychopathological.3 Clearly, one treatment program cannot adequately address the needs of all violent offenders, given their heterogeneity. Specifically, predatory offenders (e.g., armed robbers), are considered to have deficits in terms of competence, schema, and self-regulation, but not arousal and anxiety. Therefore, gains in managing arousal and anxiety should not be expected to result in the inhibition of violence for these offenders. That is, according to this model, anger-based intervention for predatory offenders should fail to yield reductions in violent recidivism because arousal and anxiety are unrelated to their use of violence. Such a conceptual model, however, requires validation.

Predominantly, the treatment of violent offenders has focused on anger control. This approach views violence as resulting from an offender’s inability to identify and manage anger. More recently, measures of treatment readiness, cognitive style in the form of hostile attribution biases, social problem solving, and relapse prevention knowledge and skills have been incorporated into violent offender assessments.4

Treatment intensity

The strategy for determining treatment intensity for violent offenders is not well defined. Treatment intensity must balance frequency and duration of sessions, and program integrity. Clinicians’ resilience and mental health must also be considered in determining treatment intensity because violent offenders are a challenging group. The setting in which treatment is provided also complicates the issue of the intensity of a program, as it is far more difficult to provide higher intensity programs in the community than in institutions or residential programs.

The range and severity of the treatment needs, then, not criminal convictions should determine the ideal length of a program for violent offenders. Currently the range in duration of such programs is 4 to 6 months with a minimum of 135 hours programming, although some programs provide 240 hours of combined group and individual treatment.

Residential versus community-based programs

Poor attendance is a key issue in community-based intervention. Demanding more intensive treatment, then, would be problematic in the community. Residential programs provide increased control to clinicians and compliance is higher, although attendance and punctuality are far from perfect. Also, programming can be more flexible, (e.g., longer programs are more easily accommodated, as are programs that require more frequent sessions or morning or afternoon sessions). Lastly, institutional programs increasingly seem to focus on identifying and coping with high-risk situations as an important aspect of treatment.

Treatment programs

It should be clear thus far that interventions for chronic or persistently violent offenders must be multi-modal and multi-faceted. A review of tertiary level pharmacological and psychological programs follows.

Pharmacological interventions

At this time, no medication has been developed or approved specifically for the treatment of violent behaviour. Several classes of psychotropic medications, however, have been utilized with some success with specific types of violent individuals. While positive reports about the impact of medication on violent behaviour are encouraging,

this literature is plagued with numerous methodological problems, including small sample sizes, lack of control groups, failure to utilize double-blind procedures, issues of non-compliance, and poor diagnostic accuracy. As well, although medication may have an impact on certain biological causes of violent behaviour, on its own it is rarely effective in reducing violence over the long term.

Psychological interventions

Psychotherapy and social casework have not proved effective at reducing antisocial behaviour.5 In the juvenile literature, multidimensional programs such as those involving family systems have had the greatest impact. Implicit in the proliferation of anger control programs is that violent offenders are angry and that their level of anger exceeds that of non violent offenders. Accordingly, reduced levels of anger are anticipated to result in less frequent and optimally less violent behaviour. This is a curious notion in that violence is relatively infrequent, unreliably measured, and often appears to be motivated for reasons other than anger.6 Recent programs now include skills practice in the areas of social skills, assertion, problem solving, and empathy.

The Rational Behaviour Therapy approach more specifically emphasizes the role of cognitions, notably irrational beliefs, in the provocation and maintenance of anger levels. Offenders are taught that their irrational beliefs result in increased arousal (anger) and that their arousal precipitates aggressive behaviour. Intervention targets the link between thoughts and feelings, challenging offenders to refute irrational beliefs, presumably decreasing the likelihood of aggressive responses.

Program effectiveness

Several studies have examined the efficacy of cognitive-behavioural interventions for aggressive adult offenders. Hunter7 offered a 10-week anger management program to 28 incarcerated male offenders who had a propensity for interpersonal violence, using a control group of 27 inmates. The intervention included relaxation therapy, stress management, conflict resolution, and cognitive therapy, the latter targeting errors in thinking (hostile and aggressive thoughts), irrational beliefs, and negative self-talk. She found that treated offenders showed significant gains relative to non-treated offenders across self-report and behavioural ratings.

Hughes8 provided a 12-week anger management program to 52 incarcerated adult offenders and attempted to compare them to a control group of 27 offenders. The latter were men who either dropped out of the program after one or two sessions, or who opted not to participate in the program for a variety of reasons. The program, described as both educational and experiential, consisted of relaxation therapy, assertiveness training, moral reasoning, problem-solving, and rational emotive therapy. Hughes found that treated offenders reported post-treatment gains regarding anger scores, irrational beliefs, and in role-plays. However, there was no difference in recidivism rates between the treated and non-treated groups.

Kennedy9 compared the relative efficacy of stress inoculation treatment to a behavioral skills treatment with a sample of 37 incarcerated adult offenders. Offenders completed several self-report measures both pre- and post-treatment. As well, Kennedy completed pre- and post-treatment behavioral ratings of structured role-plays, and reviewed offender files for relevant incident reports. She found that offenders showed post-treatment gains on several of the measures. However, she also completed an interim assessment of treatment gain and found that order of presentation of treatment had no effect. The greatest treatment gain occurred in the initial phase of treatment regardless of which treatment was offered initially.

Guerra and Slaby’s10 intervention consisted of 120 aggressive adolescents, equally divided by gender, being randomly assigned to a 12-week cognitive mediation training, attention control, or no-treatment control. Pre- and post-treatment assessment incorporated measures of social cognition (beliefs about aggression), behaviour ratings, and self-report. Post-treatment gains for the treatment group were noted in terms of increased skills in solving social problems, reduced support of aggressive beliefs, and reduced aggressive behaviours (based on blind raters). The follow-up period was 24 months for the recidivism analyses. The inference is that these socio-cognitive factors regulate aggressive behaviour, yet recidivism rates for the treated subjects, although reduced, were not significantly lower than the controls.

The Correctional Service of Canada has begun the evaluation of an Anger and Emotions Management program. Recidivism data for a matched sample (on risk, age and major admitting offence) of 110 male offenders who completed the Anger and Emotions Management Program indicate it was effective. Greatest effects were noted for higher-risk offenders, with a 69% reduction in non-violent recidivism and 86% reduction in violent recidivism, although the two groups differed with respect to time at risk.11 Further, change scores on several self-report measures were significantly related to outcome. Subsequent analyses12 have indicated that treatment dropouts have violent failure rates 8 times that of the treatment group (40% versus 5%) and twice that of the controls (40% versus 17%). A newly created program performance factor was significantly correlated with recidivism (r = .32, p < .01), and approached statistical significance in a regression analysis. Finally, a comparison of 41 matched (age, risk, past program performance) pairs of offenders indicated that the controls had rates of recidivism three times that of the treated group, but this difference was not statistically significant.

Lastly, in 1996 the Correctional Service of Canada developed an intensive treatment demonstration program for incarcerated persistently violent adult offenders.13 The treatment program is intense, involving four group sessions and one individual session per week for 16 weeks. Treatment is provided by two staff a doctoral level registered psychologist and a bachelor’s level therapist. Based on a review of the literature, treatment targets include motivation for treatment and behaviour change, aggressive beliefs, cognitive distortions, arousal management, impulsivity, conflict resolution, problem-solving, assertiveness, empathy enhancement, and relapse prevention. An exhaustive multi-method assessment protocol has been developed and preliminary data are available14 that support modest gains, as measured by the test battery and behavioural ratings with more detailed analyses in terms of outcome to be forthcoming. The program received accreditation by an external panel in 1999. The conceptual framework for this program has also been adapted for implementation in a large number of sites within the Service under the auspices of a Violence Prevention Program.15

Future directions

Notwithstanding the concern about violent offenders, there exists a surprisingly small body of literature describing effective treatment efforts, particularly in contrast to other groups such as sexual offenders and spousal abusers. Most published studies do report treatment gains, but this has mainly been restricted to self-reports and has not generalized to improved recidivism rates. To date, measurement of treatment efficacy has been confounded by this over-reliance on self-report questionnaires, the absence of control groups, and problems in the definition of violent offenders.

Implicit in the proliferation of programs for violent offenders is that this will lead to reductions in violent recidivism. The evidence across programs is encouraging but not compelling. Nonetheless, offenders who complete programs appear to be more likely to succeed. The most impressive studies regarding both methodology and outcome are from the juvenile literature and reflect comprehensive multisystemic programs. Efforts should be initiated to better incorporate best practices from the juvenile literature into treatment programs for violent adult offenders. The juvenile literature also places greater emphasis on skill acquisition in the areas of family dynamics and problem solving as compared to the emphasis with adults on arousal management, although this appears to be changing. Conceptual models, then, that integrate arousal level, self-regulation, and cognitive style may prove helpful as clinicians strive to provide programs for an array of different types of violent offenders. This appears to be the direction im which the field is moving as various correctional jurisdictions de-emphasize arousal-based anger control programs or augment the range of available programs.

What are the implications for incorporating treatment into risk management strategies for violent offenders? In those programs that focus on relapse prevention, the offense cycle provides a mechanism to discover antecedents or proximal factors to an offender’s use of violence. Also, in those programs that utilize comprehensive risk appraisals, treatment provides an opportunity to comment on the intensity and nature of community aftercare and supervision. Explicit decision rules to assist clinicians against unbridled optimism might be advantageous in incorporating treatment performance into risk management strategies.

Lastly, there is increasing consensus regarding the “correct” components for a treatment program, methods to address treatment resistance, and methodology to demonstrate treatment gain and treatment effectiveness. Equally importantly, these are increasingly being applied to the specific target of violent offending.


1. 340 Laurier Avenue West, Ottawa, Ontario K1A 0P9.

2. Serin, R. C. (1994). Treating violent offenders: A review of current practices. Research Report R-38. Ottawa, ON: Correctional Service of Canada.

3. Tolan, P. and Guerra, N. (1994). What works in reducing adolescent violence: An empirical review of the field. Colorado: The Centre for the study and prevention of violence, Institute for behavioral sciences, University of Colorado, Boulder.

4.Serin, R. C. and Preston, D.L. (in press). “Managing and treating violent offenders.” In J.B. Ashford, B.D. Sales and W. Reid (Eds.)
Treating adult and juvenile offenders with special needs. Washington, DC: American Psychological Association.

5.Kazdin, A. E. (1993). “Treatment of conduct disorder: Progress and directions in psychotherapy research”. Development and Psychopathology, 5, p. 277-310. See also Quinsey, V. L., Harris, G. T., Rice, M. E. and Cormier, C. A. (1998). Violent offenders: Appraising and managing risk. Washington, DC: American Psychological Association.

6. Henderson, M. (1984). “Prison inmates’ explanations for interpersonal violence: Accounts and attributions”. Journal of Consulting and Clinical Psychology, 52, p. 789-794.

7. Hunter, D. (1993). “Anger management in the prison: An evaluation”. Forum on Corrections Research, 5, p. 3-5.

8. Hughes, G. V. (1993). “Anger management program outcomes”. Forum on Corrections Research, 5, 1, p. 5-9.

9. Kennedy, S. M. (1990). Anger management training with adult prisoners. Unpublished manuscript.

10. Guerra, N. G. and Slaby, R. G. (1990). “Cognitive mediators of aggression in adolescent offenders: Intervention”. Developmental Psychology, 26, p. 269-277.

11. Dowden, C. Blanchette, K. and Serin, R. C. (1999). Anger management programming for federal male inmates: An effective intervention. Research Report R-82. Ottawa, ON: Correctional Service of Canada.

12. Dowden, C., Serin, R. C. and Blanchette, K (in press). A follow-up evaluation of the CSC anger management program for federal male inmate:
Dropouts.

13. Serin, R. C. (1995). Persistently violent (non-sexual) offenders: A program proposal. Research Report R-42. Ottawa, ON: Correctional Service of Canada.

14. Preston, D.L. and Serin, R. C. (1999). Case file Persistently Violent (non-sexual) Offender Treatment Program. Ottawa, ON: Correctional Service of Canada.

15. Bettman, M.D., Yazar, R. and Rove, R. (1998). Violence Prevention Program Manual. Ottawa, ON: Correctional Service of Canada.