Follow by Email

"Smart people (like smart lawyers) can come up with very good explanations for mistaken points of view."

- Richard P. Feynman, Physicist

"There is a danger in clarity, the danger of over looking the subtleties of truth."

-Alfred North Whitehead

October 25, 2010

How to Increase the Effectiveness of Treatment

In an earlier blog entry, I discussed research that showed cognitive behavioral therapy (CBT) to be effective in reducing criminal recidivism. Cognitive Behavioral Treatment Reduces Recidivism I will discuss factors impacting the effectiveness of CBT in this entry. If you follow the literature on the effectiveness of various treatment modalities for criminality, you will see that some studies show that a treatment modality is effective and some studies show that the same treatment modality is ineffective in reducing further criminal acts. (Occasionally, one sees that a treatment actually causes additional criminality.)

One of the reasons for these differences is that treatment modalities for criminality are not homogeneous like for example, a specific medication. If researchers study the effect of 81 mg of aspirin, we can be relatively assured that the treatment group received 81 mg of acetylsalicylic acid in each and every study. However, researchers testing, for example, CBT may be testing a different technique of CBT, a different topic for the CBT such as domestic violence, sex offending, or general criminality, and these programs will be delivered by different service providers at differing skill levels. Each CBT program is in many ways sui generis.

A recent study took a closer look at this issue as it relates to CBT.[i] First, these researchers performed a meta-analysis on 58 different studies of CBT. This research showed that CBT resulted in a reduction of the mean recidivism rate by about 25%.[ii] The researchers also determined that the most effect CBT programs were twice as effective as the average program resulting in a decrease in recidivism of about 50% over those individuals who were not treated.[iii]

The researchers then attempted to ascertain what factors made a CBT program more or less effective. The researchers made these conclusions:

1. The most important factor in making CBT effective was high quality implementation, which was associated with low treatment dropout, close monitoring of program quality and implementation, and adequately trained CBT providers.

2. There were no significant benefits of using “brand-name” CBT programs and that it was the general technique of CBT that was effective.

3. An anger control component and an interpersonal problem solving and peer pressure component within a CBT program enhance its effectiveness.

4. A victim impact component (getting offenders to consider the impact of their behavior on victims) and a behavioral modification component (behavioral contracts and rewards/punishment schemes) appear to reduce CBT’s effectiveness.

5. That treatment effects were greater for higher risk offenders than for low risk offenders. (Refuting the hypothesis that higher risk offenders are not amenable to treatment, and confirming the hypothesis that offenders need to first be assessed for risk, with the high risk offenders receiving the treatment)[iv]

6. CBT was as effective for juveniles as for adults.

7. The treatment setting had no effect on effectiveness. Treatment in prison was as effect as treatment in the community.[v]



[i] Lipsey, Mark W., Nana A. Landenberger, Sandra J. Wilson (2007), “Effects of Cognitive Behavioral Programs for Criminal Offenders”, Campbell Systematic Reviews, The Campbell Collection, pp. 1-27.

[ii] Ibid. p 12.

[iii] Ibid p. 21

[iv] The treatment of high risk offenders, because of their otherwise higher recidivism rates, has a greater potential for reducing recidivism compared to low risk offenders who may never recidivate even without treatment.

[v] Ibid. p. 22-23.

The views expressed in this blog are solely the views of the author(s) and do not represent the views of any other public official or organization.

October 18, 2010

DV Treatment Efficacy Questioned

In my last blog entry, I discussed the dangers of potential “side-effects” of treating people convicted of crimes. The consideration of “side-effects” is especially important in domestic violence (DV) cases.

First, with domestic crimes, the victim and perpetrator are not independent of each other. The victim may continue to live with the perpetrator, or may at least intend to again live with the perpetrator. The victim and the perpetrator may remain a single economic unit, sharing income and expenses. Financial costs of interventions will not only fall upon the perpetrator, but also fall on the victim. We all know of situations where defendants lose their jobs because of incarceration or court appearances, and the victim of the crime, who is also dependant on the income from the defendant, suffers. If perpetrators are required to pay for probation and therapy, that money also comes out of the household of the victim.

Further, if the treatment, including jail time, probation, and counseling, antagonizes the defendant, the victim may experience the result of this antagonism, through verbal and even additional physical abuse. Even if the victim does not continue living with the perpetrator, if the victim has children with the perpetrator, she may be required to continue to interact regularly with him and be subject to his abuse.

Research shows that the most informing predictor of whether a domestic violence victim returns to an abusing spouse is whether the perpetrator received counseling.[i] Victims believe that the counseling will result in a reduction of the abuse, and rely on that belief.

Unfortunately, the research on the efficacy of domestic violence treatment is mixed at best. Researchers have voiced a concern: “If treatment is essentially ineffective in decreasing recidivism, then continuing to mandate treatment may be inadvertently providing these victims with a false sense of security that, in the end, may lead to a higher likelihood of future injury.”[ii]

I have examined the research on domestic violence treatment and will present it here. I use analytic and systematic literature reviews and meta-analyses as these research techniques combine multiple smaller studies into a larger study in an attempt to increase accuracy of estimates of effectiveness.

A 2008 meta-analysis identified fifty-seven studies of domestic violence counseling studies.[iii] The researchers identified four experimental studies and six quasi-experimental studies that met their predetermined eligibility criteria for research rigor. The studies involved evaluations of psycho-educational or cognitive behavioral approaches to treatment. The treatment ranged from a minimum of 8 two-hour sessions to the maximum of 32 sessions.[iv]

These researchers found an average reduction in the recidivism rates of 26%, which was statistically significant, but with great variation among the studies. These researchers were concerned that official arrest rates did not accurately reflect the amount of repeat domestic violence actually experienced by the victims as reported domestic violence is a small fraction of the actual domestic violence that occurs.[v]

The researchers looked at seven studies that included information regarding victims’ reports of continued domestic abuse. The result of their research, using victims’ reports as a measure of recidivism, showed that the treatment had no effect on recidivism. These researchers concluded as follows:

“The findings from this meta-analysis combined with the caveats above raise questions as to the value of these programs. While additional research is needed, the meta-analysis does not offer strong support that court-mandating treatment to misdemeanor domestic violence offenders reduces the likelihood of further re-assault.”[vi]

A 2006 systematic review of the literature conducted by researchers in the State of Washington reviewed nine rigorous evaluations of educative/cognitive behavioral treatment of domestic violence offenders. They concluded: “Based on our review of nine rigorous evaluations, domestic violence treatment programs have yet, on average, to demonstrate reductions in recidivism.[vii]

Another 2006 systematic review of six studies of feminist interventions (involving education regarding sexist attitudes) and six studies of cognitive behavioral interventions was completed in Maryland. The investigators concluded: “Using the Maryland criteria, none of the interventions examined in this analysis show strong evidence that they work to reduce domestic violence. Neither of the two types of treatment programs, feminist or cognitive-behavioral, produced two studies with clear significant results favoring the treatment group over the control group.”[viii] However, the researchers concluded that both the feminist and cognitive behavior interventions were classified as promising as the results of the studies showed a positive effect of the interventions in reducing recidivism, however the reductions were not statistically significant.[ix]

In most of the reported studies, the researchers reported the percent reduction in the recidivism rate without reporting the actual recidivism rate. Feder et al (2002) did report that 24% of the offenders in both the treatment group and control group were rearrested during the one year of probation following conviction. [x]

Feder et al (2002) raised an interesting point that although domestic abuse counseling itself did not appear to be effective, perpetrators failing to complete the counseling sessions had constellations of personality traits that also made them more likely to re-offend. Therefore, failure to attend the counseling sessions was a marker that could be used to identify those perpetrators who were more likely to re-offend. Victims apparently can rely on the successful completion of counseling as being an indicator of their future safety from the perpetrator.



[i] Gondolf, E. 1987, “Seeing through smoke and mirrors, a guide to batters program evaluations”, Journal of Interpersonal Violence, 12, 83-98.

[ii] Feder, Lynetter and Laura Dugan, 2002, “A test of the efficacy of court-mandated counseling for domestic violence offenders: The Broward Experiment”, Justice Quarterly, Vol. 19, No. 2.

[iii] Feder, Lynette, David B. Wilson, and Sabrina Austin, 2008, “Court-Mandated Interventions for Individuals Convicted of Domestic Violence”, Campbell Systematic Reviews, 2008:12. Pp 1-46

[iv] Ibid p. 11

[v] Ibid. p. 14

[vi] Ibid. p. 18.

[vii] Aos, Steve, Marna Miller, and Elizabeth Drake, (2006). “Evidence-Based Adult Corrections Programs: What Works and What Does Not.” Olympia:Washington State Institute for Public Policy. P. 5.

[viii] MacKenzie, Doris Layton (2006), What works in Corrections-Reducing the Criminal Activities of Offenders and Delinquents, New York, Cambridge University Press, p . 212

[ix] Ibid. p. 212.

[x] Feder, Lynetter and Laura Dugan, 2002, “A test of the efficacy of court-mandated counseling for domestic violence offenders: The Broward Experiment”, Justice Quarterly, Vol. 19, No. 2, p. 366.

The views expressed in this blog are solely the views of the author(s) and do not represent the views of any other public official or organization.

October 11, 2010

First Do No Harm

We all know that part of the Hippocratic Oath of medical practitioners includes the promise to “first do no harm”. We also know of many drugs used to effectively treat various maladies were later taken off the market because of detrimental side-effects. We expect drug companies to research the side-effects of any drugs before they are widely used on humans. Do practitioners in the people “rehabilitation” business concern themselves with potential side effects and risks of the treatment they prescribe? The answer is not usually.

In an interesting article, “Cures that Harm: Unanticipated Outcomes of Crime Prevention Programs”, researcher Joan McCord argues that evaluating programs that attempt to reduce crime for effectiveness alone, is insufficient as some treatments cause other harm—sometimes even when they are effective in reducing recidivism.[i] McCord states: “Unless social programs are evaluated for potential harm as well as benefit, safety as well as efficacy, the choice of which social programs to use will remain a dangerous guess.” She continues, “Reluctance to recognize that good intentions can result in harm can be found in biased investigating and reporting. Many investigators fail to ask whether an intervention has had adverse effects, and many research summaries lack systematic reporting of such effects.”[ii]

McCord describes several well-designed, carefully implemented studies that resulted in harmful side-effects. The first was the Cambridge-Sommerville Youth Study, which “was a carefully designed, adequately funded, and well-executed intervention program.”[iii] That study was based on the theory that criminal conduct was related to the family in which the person grew-up. In the treatment homes, a social worker visited the family, sometimes once each week, providing friendly guidance for the children and family including referring the children to needed specialists. The control homes were identical to the treatment homes, but did not receive the treatment. All program participants reported that they thought the program had a very positive effect on them.[iv]

McCord followed these children for about 35 to 40 years. Her results showed that those children who were in the program, as adults, were more likely to have been convicted of a crime. She ultimately isolated the one factor, multiple attendance at summer camp, that increased the odds of a child being convicted of a crime as an adult by a factor of ten.[v] McCord applauds this study as having been designed properly from the start with an appropriate control group to allow researchers to discern the effects of the treatment.[vi]

McCord discusses another well-thought out and well-designed program that resulted in harm called “Volunteers in Probation”. In this program juvenile delinquents were assigned a volunteer who provided tutoring services for the youth. The program participants committed more crimes than the control group. The evaluator of that program wrote:

“To those who may feel that other such programs, perhaps their own, are so much superior or so different from this program and that our findings and recommendations are irrelevant to them, we urge caution. The staff responsible for this program has reasons good enough for them to feel that their program was effective when this study began, and without this study might still have no reason to feel otherwise. If there is anything that such a study as this one demonstrates, it is the danger of relying exclusively on faith in good works in the absences of systematic data”. [vii]

McCord discusses other programs designed to reduce criminal recidivism rates that ultimate turned out to increase them. One of these programs was the “Scared Straight” program where juvenile delinquents were exposed to tough prison inmates in an attempt to scare the delinquents into becoming law abiding. The program resulted in an increase in criminal activity for those who participated in the program. It was opined that the juveniles committed more crimes to prove that they were not scared. [viii]

Another actual side-effect case, not related to program efficacy, involved the use of citizen volunteers visiting with prisoners in an attempt to change the prisoners’ anti-social thinking by interacting with and learning from volunteers who had pro-social thinking. An evaluation of the program showed that it was effective in reducing prisoners’ antisocial thinking. However, this program had a negative side-effect. The researchers found that the citizen volunteers had increased antisocial thinking. Although the prisoners were learning from the volunteers, the volunteers were also learning from the prisoners.[ix]

McCord concludes that studies which provide evidence of harmful effects are often not published as there is a strong bias against reporting adverse effects of social programs. How often do we hear someone discussing a program that they found probably didn’t work? Rarely, if ever, do we hear of negative results.

McCord writes:

“ Many people seem to be willing to believe favorable results of inadequate evaluation designs. Some accept testimonials from clients who express their appreciation of a program. Against the claim that these provide valid evidence of effect, it should be noted that each of the programs (that she describes) would have been counted as successful by this criterion. Yet the clients would have better off had they not participated in the program.”

Those of us in the people rehabilitation business are advised to keep the admonitions of “first do no harm” and “the path to hell is paved with good intentions” in mind when we embark on a new idea to reform peoples’ behaviors. While it is important to not throw our hands up in despair and falsely claim that “nothing works”, we should always be realistic, guided by sound theory and ultimately empirical research, to ensure what we do, does not result in harm to the program participants, their families, and society.



[i] McCord, Joan (2003), Cures that Harm: Unanticipated Outcomes of Crime Prevention Programs, 587 Annals of the American Academy of Political and Social Science pp 16-29.

[ii] Ibid, p. 17.

[iii] Ibid p. 17.

[iv] Ibid pp 18-23.

[v] Ibid. pp. 21-22.

[vi] Ibid. pp. 23.

[vii] Ibid. p. 24, quoting Berger, R.J. et al, 1975, Experiment in a juvenile court: A study of a program of volunteers working with juvenile probationers. Ann Arbor: Institute for Social Research, University of Michigan

[viii] Ibid. p. 26.

[ix] Andrews, D.A and James Bonta, 2010, The Psychology of Criminal Conduct, 5th Ed. New Jersey, Mathew Bender, P. 128

The views expressed in this blog are solely the views of the author(s) and do not represent the views of any other public official or organization.

October 4, 2010

Cognitive Behavioral Treatment Reduces Recidivism

Cognitive behavioral therapy has become the principal method of treating humans with behavioral issues.[i] Cognitive behavioral therapy is based on the premise that human thought processes affect behavior, that people can become more aware of their thought processes, that they can change their thought processes, and that these changes in thought processes result in changes in behavior. These programs include such programs as errors in criminal thinking. Therapy based on the cognitive-behavioral model has been shown to be effective in reducing recidivism.

MacKenzie (2006) and her colleagues conducted a literature review and a meta-analysis of the literature and concluded that “cognitive-behavioral programs appear to be effective in reducing future criminal activities of offenders.”[ii] Aos et al. (2006) and Drake et al (2009) completed a meta-analysis of 545 studies of correctional programs.[iii] They concluded that the following cognitive-behavior programs were effective in reducing recidivism (followed by the percentage reduction in the recidivism rate); Cognitive-behavior drug treatment in prison (-6.8%), general and specific cognitive-behavioral treatment programs for the general offender population (-8.2%), sex offender cognitive-behavioral treatment in prison (-14.9%), and cognitive-behavioral treatment for low risk sex offenders on probation (-31.2%). [iv] Drake et al (2009) found that cognitive-behavioral therapy in prison or the community had one of the highest cost-benefit ratios of the treatment regimes they evaluated.[v]

Andrews and Bonta (2010) argue that the majority of criminal offenders learn through doing rather than through didactic teaching, and therefore cognitive-behavioral therapies match the offenders’ learning style making it far more effective than other types of therapy.[vi]

Andrews and Bonta argue that the key to any successful treatment is to 1.) identify the risk level of offenders, and target the high and medium risk offenders for the more intensive treatment. (low risk offenders should not interact with high risk offenders), 2.) identify those criminological needs that have been associated with criminal offending, and 3.) develop a treatment program in a style and mode that reflects the individual ability and learning style of the offender.[vii] Again, Andrews and Bonta argue that programs based on the cognitive-behavioral method are the effective treatment programs to reduce recidivism.

Aos et al (2006) also found the following programs to be effective (program followed by percentage reduction in recidivism rate) : 1. drug treatment in the community (-12.6%) , drug treatment in jail (-6.0%), adult drug courts (-10.7), intensive probation supervision with treatment (-21.9%) (intensive probation supervision without treatment had no effect.); in-prison therapeutic communities with community aftercare (-6.9%), in-prison therapeutic communities without community aftercare (-5.3%).[viii]

Aos et al identified several programs that that require additional study, but may be effective as they show early promise. Those programs are therapeutic programs for mentally ill offenders, and a program called “Circles of Support and Accountability” where five volunteers visit or contact sex offenders each week.

These studies have identified many programs that don’t work. I will address those programs in another entry.



[i] MacKenzie, Doris Layton (2006), What works in Corrections-Reducing the Criminal Activities of Offenders and Delinquents, New York, Cambridge University Press, p. 112.

[ii] Ibid, p. 114.

[iii] Drake, Elizabeth K. Steve Aos and Marna G. Miller, 2009, “Evidence-Based Public Policy Options to Reduce Crime and Criminal Justice Costs: Implications in Washington State” Victims and Offenders, 4:170-196; Aos, Steve, Marna Miller, and Elizabeth Drake, (2006). “Evidence-Based Adult Corrections Programs: What Works and What Does Not.” Olympia:Washington State Institute for Public Policy.

[iv] Ibid Aos et al (2006), p. 3

[v] Ibid Drake et al (2009) p. 184.

[vi] Andrews, D.A. and James Bonta, 2010, The Psychology of Criminal Conduct 5th Ed., New Jersey, Matthew Bender, p. 539.

[vii] Ibid, pp 44-52.

[viii] Ibid Aos et al (2006) p. 3.

The views expressed in this blog are solely the views of the author(s) and do not represent the views of any other public official or organization.