Tuesday, July 23, 2013

Treatment Outcome and the Risk Principle

Two studies of interest to treatment providers are upcoming in the August edition of Sexual Abuse: A Journal of Research and Treatment.

The first is by seasoned Canadian researchers and administrators Mark Olver, Terry Nicholaichuk, Deqiang Gu, and Stephen Wong. It’s an 11-year follow-up of a large national cohort of Canadian federally incarcerated sex offenders using a brief actuarial scale based on the Static-99R. Seven hundred and thirty-two offenders who had completed treatment were compared to 107 who had not attended treatment. They found the greatest treatment effects were among moderate- and high-risk offenders. They also found that older sexual offenders (i.e., 50 or over at the time of release) re-offended at lower rates, but that there was no interaction between age and treatment effects.

Beyond offering more reasons to be cautiously optimistic in providing treatment, the authors make the important point that “entirely static tools can overestimate risk among treated offender groups, particularly moderate or high risk offenders, as reductions in risk cannot be captured by means of static scores” (p. 416). It is encouraging to see these findings in an 11-year follow-up, as it is common to see criticisms of studies as being too brief in their examinations. As a side note, they found that Corrections Canada programs for sexual offenders also produced reductions in violent re-offending.

The other study is by Melissa D. Grady, Daniel Edwards, Carrie Pettus-Davis, and Jennifer Abramson from North Carolina and Missouri. Entitled Does volunteering for sex offender treatment matter? Using propensity score analysis to understand the effects of volunteerism and treatment on recidivism, this study addresses a familiar criticism that treatment outcome studies produce results because only voluntary people participated (selection bias), therefore suggesting that treatment programs are mostly graduating those self-starters who were unlikely to re-offend in the first place. This study took place at the prison-based Project SOAR. From the abstract:

The primary finding is that offenders who volunteered for treatment did not demonstrate any differences in recidivism rates when matched with and compared to inmates who did not volunteer to participate in treatment.

However, the study also found that those who volunteered tended to have lower Static-99 scores (one third of a point, and enough to be significant). Further, findings were mixed as to the effects on violent and non-sexual, non-violent re-offense, and therefor differed from those in the Olver et al. study above.

Perhaps most interesting in the Grady et al. study is their assertion that:

(Our) findings do not provide justification to only provide treatment to those who volunteer and seek treatment. In fact, the findings indicate that clinicians who do so may be using their resources in a way that does not maximize the potential impact of treatment. Recent studies show that a focus on the highest risk offenders, consistent with the risk principle, results in substantially greater returns in risk reduc­tion… By limiting their interventions to only those who volunteer, clinicians may not be accurately targeting those individuals who could ben­efit the most from treatment.

This is an important statement. Many programs, faced with tight budgets, have skimmed only the apparent cream of their potential clients (i.e., those who request it and persuade administrators that they are good candidates for treatment). While many believe that the risk principle (which holds that the most intensive services should be allocated to those who pose the highest risk) means higher-risk sexual offenders should get deep-dish treatment, Grady and her colleagues remind us that beyond thinking about voluntary-versus-involuntary, programs should be treating those who need treatment the most.

Some implications that flow from these studies are that:

·       Treatment programs can and probably should focus on the entire individual as much as possible. Sexual offenders are often more likely re-offend violently and in non-sexual, non-violent ways.

·       It’s time to consider less those applicants who appear most likely to benefit from our services and think more about risk- and assessment-driven treatment based on the principles of risk, need, and responsivity. This can mean favoring some clients who appear even more overtly dangerous, rude, or obnoxious and finding ways to engage them in treatment.

 To these ends, it can be useful to keep the lessons from Circles of Support and Accountability (CoSA) in mind. CoSA began as an approach with particularly high-risk sexual offenders when no other approaches were possible. They would only accept the most dangerous clients, and for their efforts they have seen remarkable gains.  Perhaps it’s time to take that lesson to heart.

David S. Prescott, LICSW