Snap Shot
- The question: What do court-ordered intervention programs for people who use violence against a partner involve, how well do they reduce reoffending, what does completion or dropout mean, and what does Arizona require under A.R.S. § 13-3601.01?
- What the research shows: Meta-analyses find small average effects at best, and some of the most rigorous reviews and trials found no significant benefit. An updated meta-analysis of 59 studies found small effects that were smaller in true experiments; a Campbell review of 11 court-mandated program studies found no statistically significant benefit; another meta-analysis found significant reductions in quasi-experiments but not in randomized trials. Programs built on risk-need-responsivity principles, motivational components, or acceptance-based approaches have shown more promising short-term results in some studies.
- What it does not show: Completing a program does not show that a particular person has changed, and dropping out does not show that a particular person will reoffend, because completers and dropouts differ before treatment begins. No curriculum has been shown to be clearly superior across settings, and few studies follow participants beyond two years or examine women or same-sex couples.
- The law: A.R.S. § 13-3601.01 requires a judge to order a person convicted of a misdemeanor domestic violence offense to complete an approved treatment program; state rules set minimums of 26, 36, and 52 sessions for first, second, and third offenses. Federal probation for a first domestic violence crime requires an approved program where one is available, and military Family Advocacy Programs provide intervention after incidents meet criteria.
- Issues for each side: Prosecutors and courts can rely on programs as part of supervision but should not present completion as proof of reduced risk. Defense counsel can present participation and progress, but a certificate of completion carries limited weight in custody and sentencing decisions without evidence of change.
Programs for people who have used violence against an intimate partner, often called batterer intervention programs (BIPs) in the research and in some statutes, are among the most common responses to a domestic violence conviction in the United States. They are typically group-based, court-ordered, and delivered over several months as a condition of probation or deferred prosecution. Courts, probation officers, family courts, and military commands rely on enrollment, attendance, and completion as markers of compliance and, often implicitly, of reduced risk.
This paper describes the main program models, reviews the outcome research, explains why completion and dropout are difficult to interpret, and summarizes how programs are regulated and used in Arizona, federal, and military proceedings, with particular attention to A.R.S. § 13-3601.01. The paper uses “person who uses violence,” “participant,” or “defendant” except when a program or statute uses another term.
Program Models
Psychoeducational and cognitive-behavioral programs
The Duluth model, developed in Minnesota in the early 1980s, frames partner violence as a means of maintaining power and control and uses structured group education to challenge beliefs that support it. Cognitive-behavioral programs focus on identifying thoughts, emotions, and situations that precede violence and on building skills to respond differently. In practice many programs mix the two approaches. A survey of 238 perpetrator programs in the United States and Canada (a 20% response rate) found that programs varied in philosophy, structure, and content and in how closely they followed approaches supported by research (Cannon et al., 2016).
Newer approaches
Newer programs adapt general correctional principles or specific therapies. Risk-need-responsivity (RNR) programs match the intensity of intervention to assessed risk, target factors linked to reoffending, and adapt delivery to participants’ characteristics; researchers have argued that BIPs should incorporate these principles (Radatz & Wright, 2016). Motivational approaches add individualized planning based on motivational interviewing. Acceptance and Commitment Therapy-based programs, such as Achieving Change Through Value-Based Behavior (ACTV), focus on psychological flexibility and values (Zarling et al., 2019). Some programs address substance use or trauma alongside violence, and a small number treat couples who choose to remain together after mild to moderate violence.
What the Outcome Research Shows
Meta-analyses
The first major meta-analysis of 22 controlled studies of programs for men found effects “in the small range,” with no difference between Duluth and cognitive-behavioral models; the authors concluded that the programs added little to the effect of arrest itself (Babcock et al., 2004). An updated meta-analysis of 59 controlled studies, which included programs for men and for women, again found small effects, smaller in true experiments than in quasi-experiments when recidivism was measured by partner or police reports. In head-to-head comparisons with Duluth-model groups, some newer interventions, including Acceptance and Commitment Therapy and Circles of Peace (a restorative-justice approach), showed the largest effects (Babcock et al., 2024).
Other syntheses reach similar conclusions with different emphases. A meta-analysis limited to 17 studies with untreated comparison groups found a significant pooled effect on domestic violence recidivism in criminal justice records, which the authors described as participants being about three times less likely to recidivate (the abstract does not state the effect-size metric, so that phrase should not be read as a probability or as a ratio of recidivism rates), but no significant effect on survivor-reported violence, and the pooled effect was significant for quasi-experimental studies but not for randomized trials; 14 of the 17 studies supplied enough data for pooling (Cheng et al., 2021). The updated Campbell Collaboration review of 11 experimental and rigorous quasi-experimental evaluations of court-mandated programs (4,824 participants), which excluded studies that used dropouts as the comparison group, found a modest but nonsignificant benefit on official reports in experimental studies (odds ratio 0.79) and no difference on victim reports (odds ratio 0.99); the authors concluded that the evidence is insufficient to show that these programs are effective (Wilson et al., 2021). A meta-analysis of 31 studies found reduced recidivism with follow-up of up to two years (odds ratios of 0.52 and 0.60) but no significant effect beyond two years, and subgroup analyses favored RNR-based treatments over other approaches (Travers et al., 2021). A meta-analysis of programs for men found reductions from pre- to post-treatment, with better results when programs addressed substance use or trauma, but mixed results from the few controlled comparisons (Karakurt et al., 2019).
Individual trials
Randomized and controlled studies show what the averages conceal. In a Spanish trial, 160 men convicted of partner violence were randomly assigned to a standard 70-hour program or the same program plus an individualized motivational plan; the added component increased attendance, advanced stage of change, and reduced self-reported physical violence and assessed recidivism risk (Lila et al., 2018). In a small U.S. trial, 42 couples who chose to stay together after mild to moderate violence were randomly assigned to multicouple group or individual couple treatment and compared with nine untreated couples; men’s recidivism six months after treatment was 25% after multicouple group therapy (significantly lower than the comparison group), 43% after individual couple therapy (not significantly lower), and 66% in the comparison group (Stith et al., 2004). The study was limited to mild to moderate violence, and couples treatment is generally considered inappropriate where one partner fears the other. An evaluation of ACTV reported more favorable recidivism outcomes than existing Duluth and cognitive-behavioral programming (Zarling et al., 2019), but other researchers argued that design limitations leave its superiority in question (Gondolf et al., 2019). That exchange illustrates a general point: promising results from single evaluations need replication before they are treated as established.
Why results are hard to interpret
Several features of this literature limit firm conclusions. Random assignment of convicted defendants is difficult, and quasi-experiments often compare participants with people who differ in risk. Official records miss unreported violence, while partner reports depend on locating and interviewing partners, many of whom have separated. Follow-up periods are usually short. Programs labeled with the same model may differ in content, length, and quality. And most participants are under court supervision, so program effects are difficult to separate from the effects of monitoring.
Completion, Dropout, and What They Mean
Dropout is common, and completers differ from dropouts before treatment begins. A meta-analysis of attrition from programs for men found that dropout was associated with characteristics such as younger age, unemployment, lower income and education, and prior criminal history (Jewell & Wormith, 2010). Because the same characteristics predict reoffending, comparing completers with dropouts overstates program effects; for that reason, the Campbell review excluded studies that used dropouts as the comparison group (Wilson et al., 2021). Completion is therefore a meaningful measure of compliance, and dropout a meaningful warning sign at the group level, but neither by itself shows whether a particular participant has changed. Evidence that bears more directly on change includes the participant’s account of the violence and its effects, consistency between that account and the record, behavior during supervision, and information from partners where safe and available.
Sex, Gender, and Relationship Configuration
Most outcome studies sampled men convicted of violence against female partners, and many state standards were written with that population in mind. Women are also ordered into programs, and a review of research on women who use partner violence recommended treatment approaches developed for them rather than programs designed for men (Carney et al., 2007). The updated meta-analysis included programs for women but did not establish separate effects for them (Babcock et al., 2024). Research on participants in same-sex relationships is minimal. These gaps mean that conclusions about program effects apply most directly to men with female partners.
Limits of the Evidence
The research base consists of a modest number of randomized trials, many quasi-experiments with selection problems, and short follow-ups. Outcome measures disagree: official records and partner reports sometimes point in different directions. Few studies examine women, same-sex couples, culturally specific programs, or long-term outcomes. The research supports the conclusions that average effects of standard programs are small, that some newer and more individualized approaches are promising, and that programs function within a broader system of monitoring and sanctions. It does not support the claim that completion demonstrates reduced risk, that dropout proves future violence, or that any single model is clearly superior.
Questions to Ask About This Research
When a program is described as effective because participants had lower recidivism in criminal justice records than untreated comparison groups (Cheng et al., 2021), ask whether that conclusion follows from the design, since the same analysis found no significant effect in randomized trials or on survivor reports, and the comparison was with untreated groups rather than with other programs. When completion is offered as evidence that a defendant poses less risk, ask what exactly was counted: the Campbell review, which excluded studies comparing completers with dropouts, found a nonsignificant odds ratio of 0.79 on official reports in experimental studies (Wilson et al., 2021). When a newer model is presented as superior, ask how large the effect is and whether it has been replicated, given the debate over the ACTV evaluation (Zarling et al., 2019; Gondolf et al., 2019). When program research is applied to a woman or to a same-sex couple, ask whether it applies to this person, since most trials sampled men with female partners (Babcock et al., 2004; Wilson et al., 2021). These questions are explained in “Reading Research Claims: A Guide for Attorneys” on this site.
Intervention Programs and the Law
Federal and Military Standards
Under 18 U.S.C. § 3563(a)(4), a federal court imposing probation on a defendant convicted for the first time of a domestic violence crime must require attendance at an offender rehabilitation program approved by the court, in consultation with a state domestic violence coalition or other experts, if an approved program is readily available within 50 miles of the defendant’s residence. In federal sentencing, information about treatment participation and amenability is presented under 18 U.S.C. § 3553(a); the 2025 Guidelines Amendment 836 deleted most departures and compiled them in Appendix B, described as outcome neutral, with the same facts available for variances under § 3553(a).
In the military, domestic abuse incidents involving service members are reviewed through the Family Advocacy Program, whose Incident Determination Committee decides whether an incident meets criteria for abuse; the program provides assessment and intervention, and commanders may direct participation (Department of Defense Instruction 6400.06; see The Military Family Violence Process: FAP, IDC, and Military Protective Orders). At a court-martial, opinion evidence about an accused’s “rehabilitative potential” is governed by Rule for Courts-Martial 1001(b)(5), which requires that the opinion rest on sufficient knowledge of the accused and not principally on the severity of the offense. Expert testimony about treatment is evaluated under Federal Rule of Evidence 702, amended December 1, 2023, and Military Rule of Evidence 702; the 2023 federal amendment took effect in the military rule in 2025 (Joint Service Committee on Military Justice, 2025), and practitioners should confirm current text. Kumho Tire Co. v. Carmichael (1999) extends gatekeeping to all expert testimony.
Arizona
A.R.S. § 13-3601.01(A) provides that the judge shall order a person convicted of a misdemeanor domestic violence offense to complete a domestic violence offender treatment program provided by a facility approved by the court under supreme court rules, by the Department of Health Services, by the U.S. Department of Veterans Affairs, or by a probation department. If the person was previously ordered to complete a program under the section, the judge must order completion again unless alternative sanctions are more appropriate. The Department of Health Services must adopt and enforce standards for program approval. Under § 13-3601.01(B), if the person has previously been convicted of a domestic violence offense within 60 months, the judge may impose supervised probation, with incarceration as a condition. The person pays the cost of treatment, the program reports attendance and completion to the court, and prior misdemeanor convictions count only for offenses committed on or after January 1, 1999. The term “domestic violence offense” takes its meaning from § 13-3601.
The Department of Health Services standards, Ariz. Admin. Code R9-20-208, require minimums of 26 sessions for a first offense, 36 for a second, and 52 for a third or subsequent offense, unless a court orders otherwise; individual sessions of at least 50 minutes and group sessions of 90 to 180 minutes; completion no sooner than three and no later than twelve months after admission unless extended; groups of no more than 15; an intake assessment covering substance use, legal, family, trauma, and behavioral health history and potential for harm; and written reports to the referring court on admission, noncompliance, discharge, and completion. Programs must emphasize personal responsibility and address domestic violence as a means of asserting power and control, may not focus exclusively or disproportionately on anger management, conflict resolution, or couples counseling, and may not require victim participation. In 2023 the Arizona Supreme Court adopted Arizona Code of Judicial Administration § 5-209 (Admin. Order No. 2023-118) to govern court approval of alternative providers; it uses the same session minimums, sets added requirements for intimate partner cases, and allows a court to adjust the number of sessions based on a validated ongoing risk assessment. In felony cases, treatment is commonly imposed as a condition of probation.
Program participation also matters in family court. Under A.R.S. § 25-403.03, as amended in 2026 (H.B. 2995, 2026 Ariz. Sess. Laws ch. 263), a court deciding whether a parent has rebutted the presumption that arises from a domestic violence finding considers the parent’s attendance at a treatment program, and the statute provides that the court may not consider a certificate of completion alone as proof of rehabilitation; the certificate must be accompanied by a records waiver and proof that the program was relevant and proportionate (§ 25-403.03(E)(5)); the court may also order intervention programs as a safety measure (see Protective Orders and Custody). Expert testimony about treatment and risk is governed by Arizona Rule of Evidence 702, which was amended in 2012 to follow the federal Daubert standard (Daubert v. Merrell Dow Pharmaceuticals, Inc., 1993; State v. Salazar-Mercado, 2014) and again effective January 1, 2024 (Order R-23-0004) to conform to the 2023 federal amendment.
Special Issues in Legal Practice
The principles described above apply with equal force to every party in a legal proceeding. The science operates the same way regardless of who is asking the questions, and the research summarized here does not by itself favor either side. What differs between the parties is not the science but the vantage point — the particular places where these issues tend to surface, and the particular errors each party is best positioned to guard against. The two summaries below identify those recurring issues. They are offered for orientation, not as litigation strategy; in every case the controlling question is what the evidence in that case will, and will not, support, and the expert’s role is to inform the trier of fact rather than to advocate for either side.
Special Issues for Prosecutors
Programs are part of a supervision system, and their value in an individual case often lies in monitoring: regular attendance, reports to the court, and an early signal when a defendant disengages. Noncompliance and dropout are legitimate grounds for court action, and the research supports treating them as warning signs at the group level. The research also supports matching intensity to risk, which any party may cite when a validated assessment bears on the number or intensity of sessions.
The limits run the other way as well. Prosecutors should avoid suggesting that program completion, or its absence, establishes how dangerous a particular defendant is, and should recognize that programs written for men with female partners may fit poorly when the defendant is a woman or the relationship is same-sex. Where the evidence suggests defensive violence by a defendant who was also victimized, a standard program may not address the circumstances that produced the conviction.
Special Issues for Defense
Defense counsel can present enrollment, attendance, and engagement as evidence of compliance and of the defendant’s effort, and can point to research supporting individualized, risk-matched, or motivational approaches when a standard program is a poor fit. In custody proceedings and at sentencing, a certificate alone says little; information about what the client learned, how behavior has changed, and how that change can be verified bears more directly on the question the court is deciding.
The same research limits what completion can show. Small average effects, the absence of significant effects in randomized trials, and the selection differences between completers and dropouts mean that completion does not establish reduced risk, and, as amended in 2026, Arizona’s custody statute reportedly treats a certificate alone as insufficient for custody purposes (verify current text); the statute and the research are separate sources. Any expert should describe treatment participation and its limits rather than offer an opinion on a party’s credibility, and opinions on rehabilitative potential in courts-martial must meet R.C.M. 1001(b)(5).
This paper is for educational purposes only and is not legal advice. The author is a psychologist, not an attorney; summaries of law are general and must be verified against current authority or with a licensed attorney.
References
Babcock, J. C., Gallagher, M. W., Richardson, A., Godfrey, D. A., Reeves, V. E., & D’Souza, J. (2024). Which battering interventions work? An updated meta-analytic review of intimate partner violence treatment outcome research. Clinical Psychology Review, 111, Article 102437. https://doi.org/10.1016/j.cpr.2024.102437
Babcock, J. C., Green, C. E., & Robie, C. (2004). Does batterers’ treatment work? A meta-analytic review of domestic violence treatment. Clinical Psychology Review, 23(8), 1023–1053. https://doi.org/10.1016/j.cpr.2002.07.001
Cannon, C., Hamel, J., Buttell, F., & Ferreira, R. J. (2016). A survey of domestic violence perpetrator programs in the United States and Canada: Findings and implications for policy and intervention. Partner Abuse, 7(3), 226–276. https://doi.org/10.1891/1946-6560.7.3.226
Carney, M., Buttell, F., & Dutton, D. (2007). Women who perpetrate intimate partner violence: A review of the literature with recommendations for treatment. Aggression and Violent Behavior, 12(1), 108–115. https://doi.org/10.1016/j.avb.2006.05.002
Cheng, S.-Y., Davis, M., Jonson-Reid, M., & Yaeger, L. (2021). Compared to what? A meta-analysis of batterer intervention studies using nontreated controls or comparisons. Trauma, Violence, & Abuse, 22(3), 496–511. https://doi.org/10.1177/1524838019865927
Gondolf, E. W., Bennett, L., & Mankowski, E. (2019). Lessons in program evaluation: The ACTV batterer program study and its claims. Violence Against Women, 25(5), NP1–NP10. https://doi.org/10.1177/1077801217741994
Jewell, L. M., & Wormith, J. S. (2010). Variables associated with attrition from domestic violence treatment programs targeting male batterers: A meta-analysis. Criminal Justice and Behavior, 37(10), 1086–1113. https://doi.org/10.1177/0093854810376815
Joint Service Committee on Military Justice. (2025). Amendments to the Military Rules of Evidence since 1984 (G. E. Maggs, Ed.; 2025 ed.). https://jsc.defense.gov/Portals/99/MRE%20Amendments%202025.pdf
Karakurt, G., Koç, E., Çetinsaya, E. E., Ayluçtarhan, Z., & Bolen, S. (2019). Meta-analysis and systematic review for the treatment of perpetrators of intimate partner violence. Neuroscience & Biobehavioral Reviews, 105, 220–230. https://doi.org/10.1016/j.neubiorev.2019.08.006
Lila, M., Gracia, E., & Catalá-Miñana, A. (2018). Individualized motivational plans in batterer intervention programs: A randomized clinical trial. Journal of Consulting and Clinical Psychology, 86(4), 309–320. https://doi.org/10.1037/ccp0000291
Radatz, D. L., & Wright, E. M. (2016). Integrating the principles of effective intervention into batterer intervention programming: The case for moving toward more evidence-based programming. Trauma, Violence, & Abuse, 17(1), 72–87. https://doi.org/10.1177/1524838014566695
Stith, S. M., Rosen, K. H., McCollum, E. E., & Thomsen, C. J. (2004). Treating intimate partner violence within intact couple relationships: Outcomes of multi-couple versus individual couple therapy. Journal of Marital and Family Therapy, 30(3), 305–318. https://doi.org/10.1111/j.1752-0606.2004.tb01242.x
Travers, Á., McDonagh, T., Cunningham, T., Armour, C., & Hansen, M. (2021). The effectiveness of interventions to prevent recidivism in perpetrators of intimate partner violence: A systematic review and meta-analysis. Clinical Psychology Review, 84, Article 101974. https://doi.org/10.1016/j.cpr.2021.101974
Wilson, D. B., Feder, L., & Olaghere, A. (2021). Court-mandated interventions for individuals convicted of domestic violence: An updated Campbell systematic review. Campbell Systematic Reviews, 17(1), Article e1151. https://doi.org/10.1002/cl2.1151
Zarling, A., Bannon, S., & Berta, M. (2019). Evaluation of acceptance and commitment therapy for domestic violence offenders. Psychology of Violence, 9(3), 257–266. https://doi.org/10.1037/vio0000097
Legal Authorities
Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993).
Kumho Tire Co. v. Carmichael, 526 U.S. 137 (1999).
State v. Salazar-Mercado, 234 Ariz. 590, 325 P.3d 996 (2014).
18 U.S.C. § 3553(a).
18 U.S.C. § 3563(a)(4).
Ariz. Admin. Code R9-20-208.
Ariz. Code of Jud. Admin. § 5-209 (adopted by Ariz. Sup. Ct. Admin. Order No. 2023-118).
Ariz. Rev. Stat. Ann. § 13-3601.
Ariz. Rev. Stat. Ann. § 13-3601.01.
Ariz. Rev. Stat. Ann. § 25-403.03 (as amended by 2026 Ariz. Sess. Laws ch. 263 (H.B. 2995)).
Ariz. Sup. Ct. Order R-23-0004 (2023) (amending Ariz. R. Evid. 702, effective Jan. 1, 2024).
Department of Defense Instruction 6400.06, DoD Coordinated Community Response to Domestic Abuse Involving DoD Military and Certain Affiliated Personnel (Dec. 15, 2021, as changed).
Rule for Courts-Martial 1001(b)(5).
U.S. Sentencing Guidelines Manual, Amendment 836 (2025) and Appendix B.
Fed. R. Evid. 702; Mil. R. Evid. 702; Ariz. R. Evid. 702.
Related topic papers: Typologies of Intimate Partner Violence · IPV Risk Assessment Instruments · Protective Orders and Custody · Bidirectional Violence and the Gender-Symmetry Debate
Related topics from Arizona Forensic Services: Courts-Martial