Can professionals change abusers? The science leans towards no.

There have been many studies published on the effectiveness of domestic violence programs used to treat men who abuse their partners.  Various researchers have performed meta-analyses on those studies to aggregate our knowledge on what we know about reforming abusive people.  Unfortunately, the research so far has been rather depressing.

This article will focus on the findings of two different meta-analyses:

  • Feder L, Wilson DB. A meta-analytic review of court-mandated batterer intervention programs: Can courts affect abusers’ behavior?. Journal of experimental Criminology. 2005 Jul 1;1(2):239-62.  (Full paper.)
  • Babcock JC, Green CE, Robie C. Does batterers’ treatment work? A meta-analytic review of domestic violence treatment. Clinical psychology review. 2004 Jan 1;23(8):1023-53.  (Full paper.)

Many abusers don’t show up for domestic violence programs

Attrition rates are quite high for domestic violence programs.  Anywhere from a fifth to four-fifths of participants drop out (even though many are mandated by courts to complete treatment), with most studies showing that roughly half of the participants drop out.

Table 1 from the Babcock et al. study summarizes attrition rates:

Study author Attrition rate (higher is worse)
Taft et al. (2001), Morrel, Elliott, Murphy, and Taft (2003), and Murphy (personal communication) 18% completed <12 weeks
Flournoy (1993) CBT 19%; psychoeducational 38%;
Harrell (1991) 20%
Edleson and Grusznski (1988) Study 3 31%
Gondolf (1997, 1998, 2000, personal communication) 32% across all sites attended less than 2 months
Chen, Bersani, Myers, and Denton (1989) 37% completed less than 7 sessions
Hawkins and Beauvais (1985) 45%
Edleson and Grusznski (1988) Study 1 47%
Waldo (1988) 50%
Leong, Coates, and Hoskins (1987) ≈ 50%
Dutton et al. (1997) 52%
Hamberger and Hastings (1988) 53%
Newell (1994) 57%
Dobash et al. (1996) 66%
Babcock and Steiner (1999) 68% completed < 28 sessions
Murphy et al. (1998)
84% (of 62 men ordered to treatment) completed <full 22 weeks
Stacey and Shupe (1984) Unknown

Violent abusers generally do not want treatment, even when they are mandated by courts to complete treatment.  While they may dangle false hope in front of their victims to make them think that they may change, their actions often speak louder than words.  They resist efforts to change, even when it is in their interest to complete the treatment program and to pretend like they’re trying.

Reports of subsequent domestic violence are low*

If treatment outcomes are measured by complaints to the police of subsequent domestic violence, then recidivism rates are low.  This is likely due to 2 factors:

  1. Most domestic violence goes unreported.  Domestic violence is typically an extension of a controlling abuser extending their control and manipulation over the victim.  When they finally start assaulting their victim, they’ve already had a great deal of control over the victim whether it’s through fear, obligation, or guilt.  Victims often seek validation from their abuser and may try to protect (enable) their abuser to “help” them change.  There are also victims who are legitimately afraid of their lives should they act against their abuser.  Many of these fears are, unfortunately, very legitimate as domestic violence can lead to murder.
  2. Abusers learn their lesson: physical abuse will get them into trouble but emotional abuse will not.  They simply learn to change their tactics to something equally damaging.  Or, they act more carefully in making sure that their victim does not report further abuse.

As far as the effectiveness of DV programs go, both meta-analyses suggest a small treatment effect *if* reported domestic violence is used as the benchmark.  The Feder and Wilson meta-analysis states:

Figure 1 indicates a general pattern of positive effects on official reports of repeat victimization in these experimental studies. […] This effect roughly represents a reduction in recidivism from 20% to 13%. However, given the small number of studies (four), there is substantial uncertainty regarding the precision of this estimate.

Victims report no effect of DV treatment programs

If questionnaires sent to the domestic violence victim are used as a benchmark of success, then the results are extremely depressing: domestic violence treatment programs don’t work.

Feder and Wilson argue:

A concern with official measures is that they may not accurately reflect the amount and severity of ongoing violence. Research consistently indicates that official reports capture only a small fraction of this abuse (Dutton 1988; Straus 1991; Tjaden and Thoennes 2000). As such, the victim is viewed as the best source for information on the offender’s continued abuse.

They conclude that there is no evidence in favour of DV treatment if victim reports are used as the benchmark:

As shown in Table 2, the mean effect size for victim reports in studies using an experimental design was near zero and was not statistically significant. The effect size for quasi-experimental studies showed a small and negative effect for treatment though this finding also was not statistically significant.

One minor problem is that somewhere around of a quarter of victims do not fill out the questionnaires sent to them, creating the possibility of biased data as the non-responders may have a more positive/negative opinion of DV treatment than those who fill out the questionnaires.  Some victims are not happy about DV treatment programs because the programs unintentionally enable and embolden the abuser.  One of the limitations of existing research is that it does not have accurate insights into what’s really happening in the relationship, especially given that both victims and abusers may lie about what’s going on.

Victims should stay away from their abusers

The flipside of the victim questionnaires is that abusers continue to be as toxic even after their arrest.  With further research, we may be able to accurately determine whether their toxicity increases or decreases slightly post-conviction.  Nonetheless, the big picture is clear: victims will continue to be victimized if they choose to go back to their abuser.  They should not try to go back to the relationship to change their abuser.

Leave a Reply