Article Text


Review: current therapies for men committing acts of domestic violence are of limited benefit
  1. Harriet L MacMillan, MD,
  2. C Nadine Wathen, PhD
  1. Offord Centre for Child Studies, McMaster University, Hamilton, Ontario,

Statistics from

OpenUrlCrossRefPubMedWeb of Science

Q Does treatment of men committing domestic acts of violence prevent recurrence and, if so, what types of treatment work best?


Embedded ImageDesign:

Systematic review with meta-analysis.

Embedded ImageData sources:

PsycINFO searched plus hand search of bibliographies and reference sections of five reviews. Additional information was obtained from study authors.

Embedded ImageStudy selection and analysis:

Inclusion criteria: experimental or quasi-experimental studies with a comparison group (randomised control groups or treatment dropouts); victim reports or police records (criminal reports) used to measure recidivism. Effect sizes using Cohen’s d were calculated. A hierarchical fixed effects model was used to assess the impacts of report type, study design, and treatment type.

Embedded ImageOutcomes:

Recidivism rates, treatment types.


Twenty two experimental (five studies) and quasi-experimental (17 studies), met inclusion criteria (44 effect sizes). The majority of treatments were Duluth/psychoeducational therapy (19 trial arms) and cognitive behavioural therapy (11 trial arms). Recidivism rates in untreated men were 21% (based on police reports) and 35% (based on partner reports). Overall, effects due to treatment were small (Cohen’s d in the range 0.2), indicating little influence on reduction in recidivism rates. There were no significant differences in recidivism rates between Duluth/psychoeducational therapy and cognitive behavioural therapy (see for table).


Programmes for men committing acts of domestic violence make a small difference to recidivism rates. However, these low rates are comparable to those obtained with correctional treatments with prisoners. The thrust should now be on developing therapies that are more effective.


Treatment design had a small influence on outcomes, with quasi-experimental designs yielding higher (non-significant) effect sizes than true experiments. There was inadequate power to assess effect due to treatment duration or follow up length. Publication bias was not found.


Despite recognition since the 1970s that exposure to intimate partner violence (IPV) is associated with major morbidity and mortality,1 particularly among women, there have been few rigorous studies or reviews that examine interventions to reduce IPV.2 This review is a notable exception: it provides a comprehensive critical overview of a central question in the IPV field: “does treating batterers work?”

The results, while sobering, are extremely important: clinicians, researchers, policy makers, and most importantly, victims of IPV, need to know that treatment of male batterers has little effect on preventing re-assault. The authors highlight a clinically relevant message when referring to the results of the five randomised controlled trials (RCTs) they review: “Based on a partner report, treated batterers have a 40% chance of being successfully nonviolent, and without treatment, men have a 35% chance of maintaining nonviolence.” It is essential for women and their caregivers, particularly those advising them about safety planning, not to overemphasise the role of batterer treatment in reducing recidivism. Unfortunately, the evidence for approaches aimed at women to prevent recurrent IPV exposure is extremely limited. Based on results of a recent systematic review,2 only one methodologically adequate intervention—post-shelter advocacy counseling—reduced re-abuse.3

Where do we go from here? Babcock and colleagues conclude that their results do not warrant abandoning existing batterer intervention programmes. We agree. However, it is imperative that existing and new programmes include high quality evaluation, the results of which should directly influence any future dissemination efforts. Too often, a programme is implemented widely, without evidence of its effectiveness (indeed, this happened with universal screening for IPV—dissemination of screening without evidence of effective interventions to help women identified as experiencing IPV).

Key gaps in the area of batterer interventions include evaluating the treatment of men who attend voluntarily, rather than being court mandated, as well as individual forms of therapy. While some argue that the ethical and legal challenges of IPV research preclude experimental studies, we assert that the trial by Dunford,4 included in the Babcock review, shows that it is not only feasible but necessary to conduct methodologically rigorous studies to determine the effectiveness of IPV interventions.


View Abstract
  • Web-only Table

    The table is available as a downloadable PDF (printer friendly file).

    If you do not have Adobe Reader installed on your computer,
    you can download this free-of-charge, please Click here


    Files in this Data Supplement:

    • [View PDF] - Table Effects sizes for recidivism rates (no significant difference between any groups)


  • For correspondence: Julia Babcock, Department of Psychology, University of Houston, Heyne Building, Houston, Texas, USA;

  • Sources of funding: none specified.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.