Are Psychological Treatments for Adults With Histories of Violent Offending Associated With Change in Dynamic Risk Factors? A Meta-Analysis of Intermediate Treatment Outcomes

• Published date: 01 December 2020
• Date: 01 December 2020
• DOI: 10.1177/0093854820956377
• Subject Matter: Articles

CRIMINAL JUSTICE AND BEHAVIOR, 2020, Vol. 47, No. 12, December 2020, 1585 –1608.

DOI: https://doi.org/10.1177/0093854820956377

Article reuse guidelines: sagepub.com/journals-permissions

© 2020 International Association for Correctional and Forensic Psychology

1585

ARE PSYCHOLOGICAL TREATMENTS FOR

ADULTS WITH HISTORIES OF VIOLENT

OFFENDING ASSOCIATED WITH CHANGE IN

DYNAMIC RISK FACTORS?

A Meta-Analysis of Intermediate Treatment Outcomes

NINA PAPALIA

BENJAMIN SPIVAK

MICHAEL DAFFERN

JAMES R. P. OGLOFF

Centre for Forensic Behavioural Science

Swinburne University of Technology

Victorian Institute of Forensic Mental Health

This meta-analysis examined whether psychological treatments delivered to adults with histories of violent offending in correctional

and forensic mental health settings were related to change in dynamic risk factors. Twenty-two controlled studies (86% comprised

male samples) were identified via a comprehensive search. Treatments had a significant effect overall, with small to moderate

improvements in trait anger, social problem solving, impulsivity, and general social skill. The treatment effect for antisocial cogni-

tions was nonsignificant. There were a small number of significant moderators, which provide preliminary indications of treatment

features that may promote greater change. However, small sample sizes and difficulties disentangling moderating effects preclude

drawing firm conclusions. While these results are useful and promising, the field remains limited by few high-quality outcome

studies, disparate outcome/measure selection, inconsistent/incomplete reporting of evaluations, and limited information about

whether change in dynamic risk factors predicts reductions in recidivism. Suggestions for advancing the field are provided.

Keywords: violent offending; treatment; meta-analysis; forensic mental health; dynamic risk

AUTHORS’ NOTE: The authors are thankful to Darcy Coulter, Cieran Harries, and Melanie Simmons for their

assistance with study selection, data extraction, and data coding. Thanks also to Shao Yuan Chong for his assis-

tance with reviewing tables and figures for completeness. The data used in this meta-analysis can be provided

to readers on request. Please contact the corresponding author for further information. All authors are affiliated

with the Centre for Forensic Behavioural Science, Swinburne University of Technology, and Victorian Institute

of Forensic Mental Health. Authors N.P., M.D., and J.R.P.O. hold appointments with the Victorian Institute of

Forensic Mental Health (Forensicare). This report was prepared by Catalyst, a consortium jointly funded by the

Department of Justice and Community Safety Victoria and the Victorian Institute of Forensic Mental Health

(Forensicare). The funders had no direct involvement in the study design, data collection, analysis, and interpre-

tation; the writing of the report; and the decision to submit the article for publication. Correspondence concern-

ing this article should be addressed to Nina Papalia, Centre for Forensic Behavioural Science, 1/582 Heidelberg

Road, Alphington, Victoria 3078, Australia; e-mail: npapalia@swin.edu.au.

956377CJBXXX10.1177/0093854820956377Criminal Justice and BehaviorPapalia et al. / Meta-Analysis of Violent Offending Dynamic Factors

review-article2020

1586 CRIMINAL JUSTICE AND BEHAVIOR

Interpersonal violence is a global public health problem, with manifest costs and conse-

quences that span significant financial strain on the criminal justice and health sectors,

adverse psychosocial effects for victims and their families, weakened public perceptions of

safety, and reduced quality of life for those who perpetrate violence (Ross et al., 2013; Serin

et al., 2009). Most violent crimes are perpetrated by a small group of persistently violent

individuals (Falk et al., 2014) who have higher rates of both violent and nonviolent recidi-

vism than other justice-involved individuals. These individuals are often imprisoned or, for

some who experience mental illness, detained in secure psychiatric facilities for the dual

purpose of protecting the community and promoting rehabilitation (Daffern et al., 2017).

Recent estimates suggest that people with convictions for violence make up a large propor-

tion (up to 70%) of imprisoned, securely hospitalized, and community-supervised individu-

als (Sturge, 2018; Völlm et al., 2018).

It is difficult, then, to overstate the value of psychological treatments for people who are

violent that lead to meaningful reductions in violent recidivism. However, theoretical and

empirical developments in psychological treatments for this population remain relatively lim-

ited (Polaschek & Collie, 2004) and high-quality evidence of the impact of treatment is surpris-

ingly small (Polaschek, 2019), particularly for violent individuals experiencing mental illness

(Morgan et al., 2012; Papalia et al., 2019). In the first meta-analytic review of interventions for

adults with violent convictions, Jolliffe and Farrington (2007) could only locate 11 rigorous

quasi-experiments and randomized controlled trials (RCTs). Only eight of these evaluations

measured the effects of intervention on violent recidivism. Subsequently, Papalia et al. (2019)

were only able to locate and meta-analyze the impact of 16 rigorous evaluations of psychologi-

cal treatments on violent recidivism. Collectively, these two meta-analyses indicate that treat-

ments for violent individuals are effective with a reduction in violent recidivism of 8 to 10

percentage points in favor of those receiving treatment. Nevertheless, this effect is modest,

highlighting the need to improve treatments for people with histories of violent offending.

Understanding the mechanisms of behavioral change is crucial to deciding the appropri-

ate focus and necessary features of treatments for those who use violence, as well as for

advancing theories of violence and aggression (Gilbert & Daffern, 2010; Polaschek, 2019).

According to the Psychology of Criminal Conduct, which underscores the risk, need, and

responsivity (RNR) principles, a reduction in the likelihood of violence occurs via a weak-

ening in the strength of dynamic risk factors (Andrews & Bonta, 2014). Several dynamic

risk factors have been elucidated, and some empirical support exists for the proposition that

reduction in the presence and relevance of dynamic risk factors is the mechanism of change

in violent individuals (e.g., Lewis et al., 2013). As such, targeting dynamic risk factors in

treatment has become the fundamental strategy in contemporary violence reduction treat-

ment programs (Daffern et al., 2017; Klepfisz et al., 2016). However, little is known about

whether dynamic risk factors actually change in violence treatments, for whom change is

most likely to occur, in what settings/contexts change occurs most readily, and which treat-

ment features are most relevant to change (e.g., treatment type, duration, delivery format,

content). Ross et al. (2013) conducted a qualitative systematic review (N = 10) of the

impact of psychological treatments on violent behavior in clinical and forensic settings.

Although outcome measures pertaining to change in dynamic risk factors (or intermediate

treatment outcomes1) were included in their review, these authors did not systematically

analyze the impact of treatments on these intermediate targets. There is meta-analytic evi-

dence to suggest that violence treatments that incorporate components addressing anger