The Violence Prevention Program in South Australia: A Recidivism and Cost–Benefit Analysis Pilot Study

• Author: Gene Mercer,Henry Pharo,Shawn Sowerbutts,Emma Ziersch,Andrew Day
• DOI: 10.1177/00938548211038333
• Published date: 01 January 2022
• Date: 01 January 2022

CRIMINAL JUSTICE AND BEHAVIOR, 2022, Vol. 49, No. 1, January 2022, 20 –36.

DOI: https://doi.org/10.1177/00938548211038333

Article reuse guidelines: sagepub.com/journals-permissions

© 2021 International Association for Correctional and Forensic Psychology

20

THE VIOLENCE PREVENTION PROGRAM IN

SOUTH AUSTRALIA

A Recidivism and Cost–Benefit Analysis Pilot Study

GENE MERCER

EMMA ZIERSCH

SHAWN SOWERBUTTS

Department for Correctional Services, South Australia

ANDREW DAY

The University of Melbourne

HENRY PHARO

Department for Correctional Services, South Australia

Rehabilitation of incarcerated men is a primary focus of correctional systems across the world. The present pilot study

examined the effect of participation in the South Australian Violence Prevention Program (VPP) on recidivism trajectories.

Individuals who participated in the VPP were significantly less likely to engage in violent recidivism, with the greatest

effect observed between Aboriginal and/or Torres Strait Islanders relative to similarly matched comparators. In addition,

the types of violent crimes committed were less severe for those who engaged in treatment relative to the comparison

group. No differences were observed between groups in overall rates of reoffending, or the length of time following

release before reoffending. The study also quantified the economic impacts of treatment and found it was associated with

a positive cost–benefit ratio of Aus$1.13. The results provide evidence that the VPP does reduce the rate of violent

recidivism, and that these results translate into economic benefits for society.

Keywords: recidivism; violence; criminogenic programs; propensity score matching; ethnicity

AUTHORS’ NOTE: The authors would like to acknowledge and thank the Department’s Executive Team for

their support in both conducting and subsequently publishing the findings of this study. The authors would

also like to acknowledge the clinicians who deliver the rehabilitation program. Their expertise and dedica-

tion to improving the lives of individuals in prison is integral to its success. This study was conducted by staff

from the Offender Development Directorate within the Department for Correctional Services South

Australia, along with support and encouragement from Andrew Day who is an Enterprise Professor in the

School of Social and Political Sciences at the University of Melbourne, Australia. All opinions expressed are

those of the authors and do not represent the opinions of the Department for Correctional Services, South

Australia or the University of Melbourne. Correspondence concerning this article should be addressed to

Gene Mercer, Department for Correctional Services, P.O. Box 1747, Adelaide, 5001, South Australia;

e-mail: gene.mercer@sa.gov.au.

1038333CJBXXX10.1177/00938548211038333Criminal Justice and BehaviorMercer et al. / The VPP in SA: A Recidivism and Cost–Benefit Study

research-article2021

Mercer et al. / THE VPP IN SA: A RECIDIVISM AND COST–BENEFIT STUDY 21

INTRODUCTION

Those who work in correctional settings in Australia and elsewhere are particularly inter-

ested in identifying ways to rehabilitate those who are serving custodial sentences for offenses

involving violence. This is not only because of the high level of harm that often results from

violence, but also because of the large proportion of sentenced people who have committed

violent offenses (nearly half of all the people in prison in Australia; Australian Government

Productivity Commission’s Report On Government Services [ROGS], 2019). In addition,

the economic costs of violence are significant, both in terms of direct costs (Smith et al.,

2014, estimated that the direct cost of homicide, assault, sexual assault and robbery was

Aus$5.4 billion in one year in Australia alone), and the burden that is placed on correctional

budgets (the cost of imprisonment in 2017/2018 was Aus$229 per day or Aus$83,524 per

year per person; Australian Government Productivity Commission, 2019). Given that a sub-

stantial proportion of the correctional budget is dedicated to managing those who return to

corrections for new offenses (nearly half of all people released from prison in Australia

receive a new sentence within 2 years; ROGS, 2019), the importance of identifying effective

rehabilitation strategies is obvious. And programs that lead to even a modest reduction in

reoffending represent a worthwhile return in terms of the harms prevented and savings to the

public purse.

A particular challenge for correctional organizations is how to identify which types of

rehabilitation programs offer the most promise. Most rehabilitative programs offered around

the world are based on the Risk-Need-Responsivity (RNR) model, first disseminated in the

1990s (Andrews et al., 1990), and include elements of social learning and cognitive behav-

ioral approaches (Auty et al., 2017). The Violence Prevention Program (VPP) is one of a

suite of programs offered by the Department for Correctional Services, South Australia.

Other programs offered address domestic and family violence and sexual violence. All of

these programs follow the RNR model. In addition, the domestic and family violence pro-

gram is informed by feminist theory (Duluth model). The VPP is offered to men whose

crimes have been generally violent. It is not offered to those whose violence had been pre-

dominantly sexual or whose violence has been toward intimate partners.

Despite the dominance of the RNR model over the past three decades, the evidence-base

pertaining to the efficacy of RNR-based rehabilitation programs for people who have com-

mitted violent offenses is surprisingly limited, especially when only methodologically

robust evaluations are considered (Higgs et al., 2019). In fact, the published studies in this

area have been characterized as providing inadequate descriptions of program content and

delivery, not utilizing untreated comparison groups, and relying on outcome measures that

do not include indices of reoffending (Polaschek & Collie, 2004). Many of the other studies

in this area have also been conducted outside of prison settings (e.g., forensic mental health;

see Papalia et al., 2019, or in community settings; see Henwood et al., 2015) and their rel-

evance to prison rehabilitation is therefore limited. Furthermore, as many of the existing

studies rely on small samples and involve programs that differ in terms of both content and

setting, efforts to aggregate these results into an overall effect size are unlikely to be mean-

ingful (see Hockenhull et al., 2015).

A review by Ross et al. (2013) identified only 10 peer-reviewed evaluation studies that

met their methodological quality criteria. Although most of these (n = 8) reported some

evidence of reduced aggression following program completion, few considered the longer-

term impact of program participation on violent reoffending. One of the most recent prison