Understanding programme drop-out from family violence programmes

What works

All of us working in the area of family violence intervention are acutely aware of the issue of attrition from programmes. I know that I often worry about the safety of partners and children of men who drop out of intervention. If we can go by general criminal justice research we know that programme drop-out has been associated by higher recidivism rates and more severity of criminal activity. So what makes the difference in terms of who drops out and who stays to complete family violence intervention? Is it the individual characteristics of the person? Is it the nature of the programme design? Is it related to men’s own experience of victimization? These are interesting questions to explore and while some aspects have been answered, others haven’t.

I was therefore pleased to read a well-constructed and astute article titled Variables associated with attrition from domestic violence treatment programs targeting male batters: A meta-analysis, by Lisa Jewel and Stephen Wormith (Criminal Justice and Behavior, vol.37, No.10, 2010). This article did not replicate the same old issues around demographic variables or fixed characteristics of an individual which we know so much about already, (e.g. age, income, education, employment status, marital status, ethnicity, etc.) but explored the relationship of men’s own victimization, programme type (feminist versus cognitive behavioral), and other intrapersonal characteristics on attrition from programmes (e.g. alcohol and drug use, depression, risk level, motivation, psychopathology and therapeutic alliance). They also explored drop-out in relation to the three stages (before assessment, between assessment and programme engagement, and during intervention) which to my knowledge is new and fertile ground.

Key results:
In terms of demographic variables:

• Employed individuals were 20% more likely to complete programmes than unemployed men
• Older men were 16% more likely to complete programmes than younger men
• Men with less education were more likely to drop out of feminist psycho-educational programmes versus cognitive behavioural programmes
• Men with more education were likely to complete as opposed to men with less education.

In terms of violence related variables:

• Referral source was most strongly associated with attrition with court mandated men 16% more likely to complete treatment than non-mandated (it is interesting to view this finding in the context of the current debates on whether to mandate or not, family court clients to programmes)
• Men attending for the first family violence incident were 14% more likely to complete treatment than men previously arrested or convicted for domestic violence.

In terms of intrapersonal variables:

• Both alcohol and drug use were associated with treatment attrition (12% and 10% respectively). It is worth noting that this is also a factor in terms of recidivism
• Cognitive functioning was also found to be a factor in attrition. Less education may be a mask for lower cognitive functioning and these men may find it more difficult to ‘keep up’ in programmes and therefore drop-out.

According to the authors the results also fit nicely in the risk, need & responsivity model developed by Andrews and Bonta (1990). This has been the core approach to general offender work for the past 20 years. It is based upon the idea that treatment intensity should be matched to level of risk posed, focussed on reducing criminogenic needs (crime causing factors) and that programmes match the ability of the person and their learning styles (including cultural matching). The overall results vindicate what this writer has been arguing for a long time; that one size does not fit all and that our programmes need to be take account of responsivity factors far more carefully. In an ideal world we would be much more careful in getting the right match of men to programme type and style. This would therefore mean better retention in programmes and the associated outcome of further enhanced safety for others.

Comments are closed.