Victim Impact Panel Research Equivocal

Several counties in Wisconsin, including my home county, Dodge, use a victim impact panel (VIP) to attempt to reduce OWI recidivism. I am aware of four studies of VIPs. The results of these studies on reducing OWI recidivism are equivocal.

At a VIP, victims of OWI homicides or injuries, recount their losses to a group of OWI offenders. The theoretical framework for VIPs is derived from concepts of inclusionary social control and re-integrative shaming coming from the restorative justice approach to controlling criminal behavior. The argument is that eliciting shame in OWI offenders from listening to victims of OWI may induce them to not engage in anti-social behavior such as operating while intoxicated.[i]

One randomized study in New Mexico showed no effect on OWI recidivism after two years.[ii] Another more recent randomized study on first-time OWI offenders, in a different county in New Mexico found no significant differences in either OWI recidivism after two years or alcohol consumption after two years between individuals who attended a victim impact panel and those who did not attend such a panel.[iii]

A nonrandomized study (individuals were not randomly selected for the VIP treatment) done again in New Mexico, showed no effect on OWI recidivism for first-time male and female and repeat male OWI offenders. The study did show that a VIP actually doubled the probability of re-arrest for female repeat offenders.[iv]

A different view of VIPs comes from another nonrandomized study in Georgia. The VIP studied involved a 60 to 90 minute program consisting of four or five victims of drunk driving discussing how drunk driving had impacted their lives. The program allowed a maximum 75 participants. This study showed that a VIP had a significant impact in reducing recidivism. After five years, those attending a VIP had OWI recidivism rate of 15.8% and a comparison group who had not attended a VIP had a 33.5% OWI recidivism rate. Almost all of this reduction occurred in the first two years, with any VIP effect on recidivism waning dramatically after two years. [v]

The research relating to VIPs is not clear. Further, future research is needed to understand content and structure of a VIP. The theoretical underpinning of a VIP suggest that a more intimate smaller group would have a greater emotional impact than a large VIP held in an auditorium setting. Further, the quality and number of the presentations made by victims may make a difference on a VIP’s impact.

As a district attorney and judge, I have attended several VIPs and I also have these concerns. The first VIP I attended was with a smaller group, and I believed the intimacy of the smaller group had a much greater emotional impact on me than the larger group(s) I attended. Had I become habituated to listening to the impact of the victims after the first time? Does this mean that repeat offenders may not be the ideal group for a VIP?

In the larger groups, I was concerned about the normalizing effect on OWI offenders from seeing so many people who had been convicted of drunk driving. Did these offenders believe that this behavior was normal behavior after seeing so many people who had been convicted of a similar offense as they had?

Further, offenders were publicly identified by raising hands as to how many were first offenders, second offenders, and so on. I was again concerned about the impact that identification had on offenders. Did those convicted of OWI 1^st think, “My behavior was not nearly as bad as those repeat offenders”?

The effectiveness of a VIP has not yet been proven with any confidence.

---

[i] Braithwaite, John, 1989, Crime Shame, and Reintegration. Cambridge, Cambridge University Press.

[ii] Polacsek, Michele, Everett M. Rogers, W. Gill Woodall, Harold Delaney, Denise Wheeler, Nagesh Rao, 2001, “MADD Victim Impact Panels and Stages of Change in Drunk Driving Prevention”, J. Stud. Alcohol 62 (3), 344-350.

[iii] Wheeler, Denise R., Everett M. Rogers, J. Scott Tonigan, W. Gill Woodall, 2004, “Effectiveness of customized Victim Impact Panels on first-time DWI offender inmates”, Accident Analysis and Prevention, 36, 29-35.

[iv] C’de Baca, Janet, Sandra Lapham, Susan Paine, and Betty Skipper, 2000, “Victim Impact Panels: Who is sentenced to attend? Does attendance affect recidivism of first-time DWI offenders?”, Alcoholism: Clinicial and Experimental Research 24: 1420-1426

[v] Rojeck, Dean G., James E. Coverdill, Stuart W. Flors, 2003, “The Effect of Victim Impact Panels on DUI Re-arrest Rates: A Five-Year Follow-up”, Criminology, (41) 4, 1319-1340

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.

Alcohol Monitoring Devices

A study evaluated the effect of the use of a SCRAM alcohol monitoring device on reducing recidivism for driving while intoxicated offenders.[i] The SCRAM ankle bracelet detects alcohol from samples of sweat collected in the air above the skin around the ankle bracelet and either stores or transmits the data. The SCRAM device is designed to thwart tampering.

On the average, offenders wore the SCRAM device for 70 days. Only 4 out of 114 SCRAM wearers were arrested for another criminal offense while wearing the device.

The researchers found that for every one year increase in age, there was an approximate 3% reduction in the probability of recidivism. They also found that for every additional prior OWI, there was an approximate 11% increase in the probability of recidivism.

The researchers also found offenders that wore the SCRAM device less than 90 days had no long-term reduction in recidivism rates compared to those who didn’t wear it (the comparison group). However, if offenders wore the SCRAM device for at least 90 days the overall recidivism rate was about one half the rate of the comparison group.

This result comports with research that suggests that 90 days of drug treatment may be the minimum length of treatment to detect any treatment effect.[ii]

The researchers warn that these are preliminary results that need to be replicated by others. However, at this time it appears that the use of a continuous alcohol monitoring device such as a SCRAM bracelet may be an economical method to reduce recidivism in OWI offenders.

---

[i]Flango, Victor E. and Fred L. Cheeseman, (2009). “Effectiveness of the SCRAM Alcohol Monitoring Device: A Preliminary Test”, Drug Court Review, Vol. VI, 2. p. 109-133.

[ii] Marlowe, D., D. DeMatteo and D. Festinger (2003). “ A sober assessment of drug courts.” Federal Sentencing Reporter. 16, 153-157

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.

Research Mixed on Efficacy of DWI Courts

Driving while intoxicated (DWI) courts have been modeled after drug courts. These treatment courts use on-going court review hearings coupled with intensive alcohol and drug treatment and random or continuous testing for the use of drugs and alcohol. Participants in these DWI courts receive negative reinforcements such as additional jail for program violations, and positive reinforcements for program achievements.

A recently published systematic literature review of 41 published and unpublished evaluations of DWI courts attempted to evaluate the effectiveness of DWI courts on reducing recidivism.[i] The researchers reviewed the methodology of these evaluations, and eliminated all but 14 for further review because of methodological limitations with the evaluations. Of the remaining fourteen, merely five of the evaluations satisfied at least 65% of these researchers’ recommended criteria for evaluating a DWI court’s effectiveness.

The authors of this literature review concluded, “Although the results hint at emerging evidence potentially favoring the effects of DWI Courts, conclusions are seriously hampered by the disappointing state of the research in this area.” [ii] These studies were unable to answer some significant questions. What component of the DWI court was the operant variable that may be impacting the recidivism rate? Was it the more comprehensive AODA treatment provided? Was it the intensive supervision and monitoring? Was it the additional required court review hearings? For which group of offenders is a DWI court effective? These questions remain unanswered, and can only be answered by additional research.

Some research has shown that drug courts may be effective. However, a major difference between alcohol abuse and drug abuse is that the use of drugs is illegal or highly regulated, whereas the use of alcohol is legal for adults, and only becomes illegal when coupled with driving. That factor complicates AODA treatment.

A 2010 study of two hybrid alcohol and drug courts in two, unnamed, upper mid-west cities compared DWI offenders who were sentenced to prison and then paroled, with DWI offenders who were enrolled in a hybrid alcohol and drug court. The researchers reported no statistical difference for re-arrest for a new DWI offense between those who completed the alcohol and drug court program and those who completed prison and then parole. (There was some evidence of a reduction for arrests for other crimes for alcohol court participants, but it did not reach statistical significance.) However, the hybrid court reduced recidivism for non-DWI offenders.

The researchers concluded “Specifically, this study determined that completion of a hybrid drug court program did not benefit the subsample of chronic DUI offenders compared to similar DUI offenders who completed prison terms followed by parole supervision (even when weighting procedures were employed to counter the effect of small sample size).”[iii]

At this point there is some research which supports the efficacy of DWI courts for reducing recidivism. DWI courts are resource intensive in terms of treatment, supervision, and court time. Further research is needed to discover whether or not each of these additional resources reduce recidivism and provide a return on investment in terms of community safety. At this time, the research is unclear on the question of DWI courts’ effectiveness.

[i] Marlowe, Douglas B., David S. Festinger, Patricia Arabia, Jason R. Croft, Nicholas S. Patapis, and Karen Dugosh, (2009), “A Systematic Review of DWI Court Program Evaluations”, Drug Court Review, Vol. VI, 2.

[ii] Ibid. p. 32

[iii] Bouffard, Jeffrey A., Katie A. Richardson, Travis Franklin, (2010) “Drug courts for DWI offenders? The effectiveness of two hybrid drug courts on DWI offenders”, Journal of Criminal Justice, 38, pp.25-33

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.

A Clash of Knowledge Cultures

The National Institute of Corrections (NIC) has been working on a project called “The Evidence-based Decision Making in Local Criminal Justice System Initiative”. Chief Justice Abrahamson, of the Wisconsin Supreme Court, is a member of the advisory board. NIC has recently solicited applications from cities and counties to become possible study sites for implementation of its framework that uses evidenced-based practices throughout the criminal justice system.

The goal of this project is to reduce criminal recidivism, and therefore reduce the number of people victimized by crime. The initiative borrows a model from the medical community that used evidence-based medical practices to successfully reduce deaths in medical patients called the “100,000 Lives Campaign”.

The framework can be accessed at: A Framework for Evidence-Based Decision Making in Local Criminal Justice Systems

Appendix Three to this document is a matrix of research findings about what works and doesn’t work to reduce pretrial misbehavior and criminal recidivism, and is worth the look by anyone involved in the criminal justice system.

This project is an ambitious attempt to use those practices that research has shown to be effective throughout the criminal justice system. The project’s leaders appear to understand the difficulty of the endeavor, and are attempting to carefully implement the framework in selected jurisdictions.

This initiative correctly recognizes differences between the medical system and the legal system and the resulting challenges due to these differences. However, I believe the authors of the initiative have missed a fundamental and critical difference between the medical and the legal community—the difference in each community’s functional theory of knowledge. I believe most lawyers view knowledge and its acquisition much differently than most physicians.

Medical training and medical practice are essentially science-based resulting in an evidence-based theory of the acquisition and creation of knowledge. Knowledge is understood to be developed through controlled empirical studies and through systematically collecting data of the results of replicated practices. Criticism is encouraged, and helps the “sifting and winnowing” of information. Physicians are trained in and habituated to problem solving using science-based principles.

Legal training and legal practice are essentially not science-based, but rather have (and to a large degree must have) an authority-based theory of knowledge. For example, the answer to a legal question is correct, not because of some empirical research has shown it to be correct, but because four members of the Wisconsin Supreme Court have said it was correct—end of story.

Lawyers and judges are accustomed to obtaining knowledge through accepting the views of those highest in the hierarchy who have addressed the issue. For example, what did the U.S. Supreme Court have to say on the issue? When the authorities have spoken clearly on the issue, there is little further argument and criticism is strongly discouraged—at least in the adjudicative forum. Dissent is not generated by the expounding of conflicting empirical studies or rational arguments from these studies, but by expounding conflicting authorities (which embody different values) and rational arguments there from. Most lawyers are not trained in scientific design and quantitative methods and are not accustomed to obtaining knowledge through critical evaluations of empirical studies.

I believe (here I go making a statement without any empirical evidence—however I am a lawyer (I can also say I have a hypothesis)), that for many legal practitioners, using the scientific method will be foreign, unfamiliar, and at times, threatening. Science has a history of appearing threatening to authoritative pronouncements—ask Galileo. Where physicians may readily accept the conclusions of published scientific studies, lawyers, because of the knowledge culture in which they are enmeshed, may hesitate. They may remain unconvinced. This epistemological difference will challenge NIC’s implementation of its “Evidenced-based Decision Making in Local Criminal Justice System Initiative”. I wish NIC well in its endeavor.

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.

Intelligent Article on Alcoholics Anonymous

Brendan I. Koerner has written an intelligent article about Alcoholics Anonymous in Wired, entitled “Secret of AA: After 75 years We Don’t Know How It Works.” The article describes recent research on the effectiveness of AA (It doesn’t appear to work when ordered by a court), but also discusses the physiology of alcohol addiction and the neurological effects of the addiction. The article further discusses some of the possible mechanisms that may be at work in AA’s success in reducing addiction to alcohol. One point made by the article, which legal practitioners must keep in mind, is that 90 percent of alcohol dependent people who had stopped drinking will drink again. If you are interested in the seeming intractability of the problems we see related to the abuse of alcohol, this article is worth your time. http://www.wired.com/magazine/2010/06/ff_alcoholics_anonymous/

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.

Jail has Minimal Effect on OWI Recidivism

Research studies evaluating jail as a judicial intervention to reduce the recidivism rate of operating while intoxicated defendants have overwhelmingly found it to be ineffective.[i] However, limited research evidence exists that a short (two day) jail sentence may result in a reduced recidivism rate.[ii]

Most of these studies were done in the 1990s. The most recent study I could find, found that jail combined with treatment was the most effective strategy among three studied. However, the jail sentences were combined with treatment, so it was not possible to discern the effect of the jail component alone. The study confirmed what other studies had found—that the length of jail did not seem to be related to offenders’ rates of recidivism.[iii] I found no research that considered the effect of prison sentences on the recidivism rate of OWI offenders.

Attempting to deter an offender from re-offending is but one reason for a sentence that involves incarceration. General deterrence of other potential offenders is another reason. Jail sentences may have an educational effect on the general public by making drunk driving less culturally acceptable.[iv] Through a jail sentence, society instills the message that drunk driving is wrong and is not an acceptable behavior.

Incapacitation is another reason for incarcerating drunk drivers. An offender is at much lower risk to re-offend if incarcerated. (However, it does happen that an offender on work-release re-offends.) With some offenders, the risk to the public from their repeated drinking and driving is so great that their treatment must be in a confined setting to protect the public.

Another reason for incarcerating drunk drivers derives not from the possible consequences of the punishment, but from the deontological view that drunk driving is deserving of punishment because it is morally wrong—period. Someone, by driving drunk, has put other individuals at risk of harm or death through negligent or reckless drinking and driving behavior. The argument goes, that the wrong itself requires retribution, regardless of any practical effect from the punishment. The deontological argument doesn’t consider behavioral changes to the defendant or others, and therefore it does not lend itself to measurement and hence empirical investigation. I will return to this topic at another time from a different angle.

The problem with incarceration is that it is not free. Society must pay for jail and prison space and their attending direct costs, as well as for the collateral costs to the families of the incarcerated. In any fiscally healthy society, the cost of government must be scrutinized to ensure tax dollars are used wisely. Every tax-dollar spent on incarceration is a tax-dollar not available for another societal need. Judges cannot escape that truth when sentencing.

Does the next dollar spent on incarceration increase community safety or the perception of justice by at least a dollar? Is a perception of justice alone enough to justify the cost of this punishment? Those are the questions that are being asked more frequently as our society struggles to address its increasing public debt. I believe they are the correct questions.

[i] Voas, Robert B and Deborah A. Fisher, 2001, “Court Procedures for Handling Intoxicated Drivers”, National Institute of Alcohol Abuse and Alcoholism.

[ii] C. Falcowski, 1984, “The Impact of Two Day jail Sentences for Drunk Drivers in Hennepin County, Minnesota, NITSA

[iii] Delaney, Harold D., et al, (2005) “Variations in Jail Sentences and the Probability of Re-Arrest for Driving While Intoxicated.” Traffic Injury Prevention, Vol. 6 Issue 2.

[iv] James L. Nichols, and J. Laurence Ross, (1990) “The Effectiveness of Legal Sanctions in Dealing with Drinking Drivers,” Alcohol, Drugs, and Driving, 6(2) 33-55.

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.

Why Aren't Drunk Drivers Deterred?

According to the research, judicial interventions result in only small reductions, if any, in the OWI recidivism rate. Of course, judicial interventions only involve individuals with a prior OWI. Judges haven’t had any prior involvement with the largest group of OWI offenders—OWI 1sts.

For that group, society must rely on simple deterrence principles alone to stop drunk driving. The classical deterrence principles of Cesare Beccaria, that are still argued every day in court rooms in Wisconsin, demand that the expected costs of drunk driving must exceed the expected value of drunk driving.

This theory is underpinned by the rational actor model of human behavior. The rational actor model of human behavior has been the foundation of economics and law for centuries. (It is almost certainly wrong, but that is a discussion for another day, and does not affect the analysis below.)

A rational actor makes the decision about whether or not to drive drunk by comparing the expected costs and the expected benefits of drunk driving. If the expected benefits exceed the costs, then the rational actor drives drunk. If not, then he finds a different way home or stops drinking before he is intoxicated. The decision point occurs either at the time the next drink will cause him to be intoxicated, or if he is intoxicated, when he decides to drive.

The expected costs can be calculated as follows: (The probability of being arrested for drunk driving multiplied by the probability of being convicted and punished if arrested multiplied by the expected punishment for drunk driving) plus (the probability of being arrested for injury or homicide by intoxicated use multiplied by the probability of being convicted and punished if arrested multiplied by the expected punishment for injury or homicide by intoxicated use) plus ( the probability that one will be injured or killed themselves and the costs of those results).

Research has shown that more than 90% of those arrested for OWI are ultimately convicted. This number has increased greatly over the last twenty five years with changes in the law regarding implied consent, forced blood draws, and the use of blood tests themselves. The punishment levels, including incarceration and fines, have increased dramatically through the years. The punishment for homicide by intoxicated use has increased many fold during my legal career.

What hasn’t changed much in Wisconsin, is the probability of being arrested for drunk driving (If an accident results, the chances of being arrested are quite high). Using the one published survey figure of one drunk driving arrest for every 55 drunk driving incidents, the chance of being arrested for drunk driving is less than 2%.

Based on the low arrest rate for drunk driving, a rational actor can conclude that one will not be arrested if he or she drives drunk. The rational actor will also correctly calculate that there is a low probability of getting into an accident. Therefore, the expected punishment for drunk driving is rationally quite low primarily because of the low probability of being arrested. The classical theory of deterrence fails because the expected costs of driving drunk are less than the expected benefits.

One attempt to change this calculus was to rescind the ban in Wisconsin on sobriety check-points. The hope was to raise the expectation of being arrested for drunk driving. The legislature, after balancing the interests as they should, decided not to lift the ban on such checkpoints.

However, it is clear to me to increase deterrence of operating while intoxicated, policies must be implemented to increase the probability of arrest for drunk driving. Changes in penalties, treatment, and other judicial interventions are all small potatoes compared to changes in the expectation of arrest.

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.