Anchorage Wellness Court: Challenges

Anchorage Wellness Court: Challenges

Ronald S. Everett

Everett, Ronald S. (Spring 2009). "Anchorage Wellness Court: Challenges." Alaska Justice Forum 26(1): 1, 5-9. This article examines the Anchorage Wellness Court (AWC), which deals with misdemeanor alcohol offenders, including those who drive under the influence (DUI); discusses the concepts of therapeutic jurisprudence and problem-solving courts; and examines the challenges to the longstanding need and demonstrated potential for success of the AWC program. The author discusses societal views of DUI drivers and explores the broader issue of "impaired" drivers: those, for example, who drive while talking on cell phones. Four recommendations are given for the AWC: to reorient and/or clarify policy goals regarding drunk drivers; to develop alternative treatment programs and expand the number of available treatment providers; to establish a 5-year follow-up program with potential for expunging arrest and conviction records and other incentives; and to develop proactive, rather than reactive, polices of prevention to confront alcohol-impaired drivers and the underlying problem of alcohol abuse and addiction. A sidebar highlights sanctions for DUIs under Alaska Statute Section 28.35.030 and Anchorage Municipal Code 9.28.020.

The Anchorage Wellness Court (AWC) deals with one of the most pervasive public health concerns in Alaska: alcohol abuse and driving under its influence. The longstanding need and demonstrated potential for success of the Anchorage Wellness Court confront many barriers that hinder its ability to address the serious and substantial alcohol problem within the community of Anchorage and throughout Alaska. The general issue of alcohol abuse and the specific problem of driving under the influence (DUI) are longstanding problems, and multiple strategies have evolved for dealing with them. Problem-solving drug treatment courts were first introduced in 1989 and alcohol courts are a recent addition.

The terms specialized courts, community courts, therapeutic courts, drug treatment courts and problem-solving courts are often used interchangeably. This essay will use problem-solving courts as the more general term and discuss the Anchorage Wellness Court as a problem-solving drug treatment court.

Problem-Solving Courts

Problem-solving courts have extended across a broad range of social problems from drug addiction and domestic violence to neighborhood social disorder. Many problem-solving courts, particularly those concerned with drug addiction, adhere to the ideas of therapeutic jurisprudence including providing positive outcomes for victims, offenders, and society; working with groups outside the justice system such as treatment providers; and using screening and assessment to identify individuals appropriate for referral to a problem-solving court. (See "Therapeutic Jurisprudence" on page 10.) Yet a number of problem-solving courts either do not acknowledge or fail to recognize the goals of therapeutic jurisprudence. For example, courts focused on domestic violence may even explicitly reject notions of therapeutic intervention, presenting control, compliance and monitoring as the primary problem-solving goals of the court rather than treatment and positive outcomes for offender, victim and society.

Therapeutic jurisprudence advocates using the law for therapeutic rather than anti-therapeutic purposes. The implications of this transformation for court operation are both practical and theoretical. Clearly it is beyond the scope of this brief essay to enumerate or discuss in any detail all of these. Nevertheless, an identified central principal of therapeutic jurisprudence requires that the court process move beyond standard notions and practices of adjudication and punishment. To achieve the crucial goal of treatment success, and to guide court action, the court must construct a more complex and meaningful explanation of the problem, taking into account the decision-making context of the individual. By applying a more complex and arguably more complete understanding of the problem, the problem-solving court is better able to implement effective treatment regimes. Individualization of a therapeutic plan based on developing a deep understanding of the problem is a central guiding principle of problem-solving courts.

Anchorage Wellness Court

Drug/Alcohol Treatment Courts

In the 1980s many criminal courts became overwhelmed with drug cases, making routine operation nearly impossible. In some jurisdictions, often those with the highest level of drug cases, many judges expressed frustration with the apparent ineffectiveness of prison and other sanctions to reduce the level of recidivism for drug offenders. Drug Treatment Courts (DTCs) emerged in the late 1980s, at least initially, as an administrative response to improve the efficiency of criminal courts. Over time, DTCs have evolved and expanded, becoming just as likely to promote therapeutic outcomes as a goal of equal importance to the increased administrative efficiency of the criminal court process. Over the past 20 years Drug Treatment Courts have proliferated throughout the U.S. to address the drug use and treatment concerns of local communities.

As a type of problem-solving court, the DTC treats an offender's drug addiction through intensive court supervision and a structured treatment program. Although there is substantial variation among Drug Treatment Courts in terms of treatment programs and overall process, most share a common underlying orientation and goals pursued within a non-adversarial therapeutic intervention.

The Anchorage Wellness Court (AWC) is a problem-solving court and conforms to the general model of a DTC or therapeutic court with a primary focus on alcohol-related misdemeanor driving offenses. Until recently, alcohol-specific courts have received less attention than DTCs, which target the needs of offenders whose drug of choice is something other than alcohol. However, with continued support from the National Highway Traffic Safety Administration (NHTSA) and pressure from advocacy groups like Mothers against Drunk Driving (MADD), these alcohol courts have become more common over the past decade. The Anchorage Wellness Court was initiated in 1999 and has since developed into a mature problem-solving court consistent with the tenets of therapeutic jurisprudence.

Treatment Program. The AWC treatment program has evolved over the life span of the court but has retained the core requirement of a substantial 18-month commitment by participants. The essential and unchanged components include abstinence from alcohol, group therapy (Alcoholics Anonymous), individual moral recognition therapy (MRT), employment or active searching for work, and weekly court attendance and monitoring. Since inception the target population of the AWC has changed and expanded and some minor elements of the treatment program have been added, altered, and/or deleted.

AWC Evaluation. A multi-year evaluation of the AWC was completed recently. This evaluation focused on the years 2000-2004, what can be considered the court's early formative period. The evaluation addressed multiple dimensions of the court process and treatment program including assessing individual outcomes and the court's overall effectiveness in reducing alcohol use and associated criminal recidivism. The evaluation also investigated the costs and benefits of the court compared to the standard criminal justice response to DUI offenders. Finally, an important, although less common, component of the evaluation was designed to gather evidence on transferability of the court process to provide guidance to others when starting a similar problem-solving therapeutic court.

Although the details of the evaluation are not the focus of this essay, a general statement on the court's overall effectiveness offers a useful context when interpreting the following discussion. Recognizing that basic methodological limitations complicate the interpretation of any outcome analyses of a non-experimental intervention, the detailed results of the evaluation will not be discussed. Nevertheless, the core findings document that offenders who choose to enter the AWC and subsequently complete the treatment program have lower recidivism rates compared to similar offenders not participating in the AWC. Keep in mind, and as we know from Table 1, that only a small percentage of all eligible offenders referred to the AWC decide to enroll. (Note: To explore the influence of participant self-selection, the outcome analyses included the estimation of propensity score models to investigate the problem of selection bias and account for the effect of self-selection into the AWC.)

Table 1. Anchorage DUI Arrests and Wellness Court Participants, 1998-2009

The complete evaluation outcome analysis also includes the results of survival models with follow-up periods of 24, 36, and 48 months. The survival models are used to examine the time to failure or continued desistance, for each participant, and the factors that may influence the process. As a result, the core finding of overall success of the AWC in reducing recidivism is based on (1) a careful and complete analysis, including efforts to assess the influence of selection bias due to the non-random self-selection of participants, and (2) that participants failed (recidivated) at varying points in time during the follow-up period and that some never failed. Although this finding is not without caveat, given the fundamental limitation of a non-experimental research design to demonstrate cause and effect, it is sufficient to justify the central conclusion of the outcome evaluation that the AWC works for those who chose to participate and complete the treatment program.

To achieve a degree of success in proportion to the enormity of the problem of alcohol abuse/addiction, the AWC must do more far more. But the AWC confronts significant barriers to expansion and greater successes. These barriers include individual characteristics, organizational program design, and law/policy issues.

Impediment to Success

Individual Barriers to Program Participation

Alcohol abuse and addiction are complex problems and often viewed as intractable. All treatment programs or efforts to facilitate desistance are complicated because alcohol use is legal and firmly embedded in our society and culture. The data in Table 1 reveals several critical issues relevant to the AWC. First we can see that arrests for DUI have been a persistent problem during the previous decade, and there appears to be a substantial increase in the past year. Additionally, comparing the number of DUI arrests to the number of alcohol-related misdemeanor offenders entering the AWC clearly indicates that only a small number of those offenders potentially eligible to participate in the AWC are deciding to do so. A nominally eligible DUI offender must request admission to the AWC and then complete a multistep admission protocol. The process begins with an expressed interest in the AWC program by the offender, followed by a review and plea offer by the prosecutor, and a brief observation by the offender of the AWC in operation. A substance abuse assessment and a review by the AWC team are next. Admission to the program is complete when the AWC judge approves the plea; the offender accepts the plea agreement, enters the AWC program, and begins the treatment plan.

This difference between the number of eligible participants and the number actually opting into the program raises important issues. Specifically for the AWC, one critical factor is determining why so few eligible offenders decide to participate. A more general question for the community to address is the decade-long trend in DUI arrests; this trend seems to raise serious doubts about the exclusive reliance on and effectiveness of deterrence-based punishment strategies. Over the past decade criminal justice punishments for DUI, and the negative collateral consequences for insurance and employment, have become more severe.

Unfortunately, we have not witnessed a corresponding decline in the number of DUI arrests over this same period of time. The assumption that there are a just a few "problem drunk drivers" who can be eliminated through increasingly severe sanctions seems to miss both the dimension and complexity of the problem. Who decides to enter the AWC to change their behavior and embrace sobriety is an obvious yet difficult question to answer. What group of offenders is most amenable and likely to benefit the most from the treatment regime provided? Do the groups overlap or are they completely different? These questions are to some degree addressed through assessment, but significant issues remain for further investigation.

These individual factors should be considered as potential barriers to expansion and success of the AWC. It seems that a deeper understanding of what "readiness for change" means in the context of the AWC is imperative. Participation in the AWC is voluntary and so documenting who chooses to enter and why is essential. Understanding the decision to participate is the necessary first step to demonstrate fully the effectiveness of the AWC absent the inherent benefits of an evaluation based on an experimental research design. The AWC will improve its effectiveness in reducing recidivism and its overall cost benefits by developing a base understanding of offender amenability and refining the target population most likely to benefit from the treatment program offered by the AWC.

Organizational and Program Barriers

Beyond some individual level of readiness to change as a prerequisite for success, participants in the AWC confront substantial program costs in time and money. The optimal length for the AWC treatment program should be carefully examined and remain the focus of continued research. The presumption in drug treatment courts has typically been that longer programs are likely better, and much related research on criminal recidivism supports this contention. More recent research has also concluded that in some contexts, brief intense treatment interventions can also be effective. However, recognizing that alcohol abusers are less inclined to make thoughtful, forward-thinking decisions of all kinds, we should concede that a treatment program that seems too long may tip those waffling on the edge of change back into their current behavior. Further complicating the court and treatment process is the inclusion of different types of alcohol offenders-ranging from serious addiction to episodic abuse to occasional alcohol-related flawed decision-making-within the AWC. This may be addressed by adjusting the incentive structure of the court to entice more offenders over the initial barriers and into treatment. Calibrating just the right dose to maximize effectiveness, cost, and participation requires ongoing assessment.

Additionally, program costs to the participant should be nominal-sufficient to induce adequate buy-in, but not so much to discourage entry or create stress over time as costs accumulate through the life of the program. To mitigate this issue most of the 14 treatment providers working with the AWC use a sliding scale to calculate participant fees. For most participants the barrier of program costs is more perception than reality, but this can be used to rationalize their decisions to not participate or drop out. Although it is difficult to calculate, and comparisons are inexact, the available data suggest that the direct participant program costs in the AWC are higher than average. Perhaps a treatment cost rebate could be offered and earned by successfully completing the AWC program and maintaining desistance for varying periods of time after graduation from the program.

Theory and Policy Barriers

At this moment in the first decade of the twenty-first century we as a community seem ready for a change in criminal justice policy direction and punishment philosophy. In many areas of criminal justice, from criminal sentencing to operation of the juvenile court, there is a reconsideration of our punishment strategies including an overreliance on prison and an intensifying exploration of intermediate sanctions. In this context it seems the moment may be here for a reappraisal of how we deal with the issue of DUI. This essay is only a beginning, offering a few suggestions to start a conversation that must continue throughout our community and ultimately among our policy makers and political leaders.

It is extremely difficult to separate out and discuss the factors that influence criminal justice policy and appropriate punishments when discussing drunk drivers. In our desire for retribution and social protection, the drunk driver has been characterized as a bad and undeserving person who merits our strongest condemnation and harshest punishment. However, in overemphasizing the drunk driver as a social pariah we have misunderstood the true nature of the problem and ultimately implemented ineffective policies. Even from the very basic data presented in this essay, it seems obvious that in talking about the problem of DUI we are talking about more than a few problem drinkers. The policy choices of the past thirty years make clear that increasing the severity of punishment can produce a measurable marginal deterrence, but it is insufficient alone to extinguish the problem.

Criminal justice policies intended to reduce DUI have implemented different strategies over the past 30 years, including education and media campaigns. The most severe are expanded detection efforts such as enhanced patrol and checkpoints and increased criminal sanctions including fines and incarceration. Many studies over the years have demonstrated either a modest deterrent impact or no demonstrable level of deterrence and reductions in DUI from these efforts. Additionally, much research indicates these interventions are costly, difficult to maintain and, most importantly, do not appear to address the underlying problem.

Alternative strategies have focused on public awareness of sanctions and also publicized society's contempt for the drunk driver. Advocacy groups like MADD have been persistent in their efforts to increase moral inhibitions and the social stigma of DUI. These strategies remain necessary and have over the past 30 years reduced the level of DUI-related accidents and deaths. Nonetheless, the persistence of the problem suggests that we must continue to search for alternative strategies as these may have reached their limit of effectiveness far short of eliminating the problem. It seems reasonable to assume that the two largest groups of offenders-the chronic alcohol addicted driver and the alcohol drinker flawed decision-maker-are best addressed with very different policies and sanctions.

Necessity of Social Protection

The public's concern about DUI often focuses on social protection and the need to feel safe when driving, an everyday routine activity for most people. The media focus on drunk drivers causing fatal accidents involving innocent others has tended to exaggerate the view of drunk drivers as a threat to public safety. Although this is a real problem to be taken seriously, it also ignores the reality that DUI is not limited to a small group of problem "killer drunk drivers." (See Tables 2 and 3.) We need to address these most serious cases, but we also need to recognize the nature of the problem. The basic descriptive data from Anchorage make it obvious that DUI is not limited to a few chronic drunk drivers, but includes a larger population of casual/social drinkers who drive after drinking, rationalizing it as a convenient necessity or perhaps honestly assessing that they are below the legal limit of blood alcohol concentration (BAC) to drive. These drunk drivers do not recognize themselves in the public images of the "killer drunk" and do not believe their behavior is a threat to public safety. They do not believe they suffer from alcoholism nor do they see themselves as alcoholic. For some, although they could perhaps benefit from the AWC, the admission of having a drinking problem is a barrier to participation.

Table 2. Anchorage Vehicle, Bicycle, Pedestrian, and Motorcycle Accidents, 2002-2007
Table 3. Alaska Fatality Analysis Reporting System (FARS) Statistcs, 1999-2008

Another category of potential AWC participant is the alcohol addicted chronic drunk driver. The alcoholic is by definition irrationally concerned with when and where to drink, and it is unlikely that the threat of sanctions by the criminal justice system is a relevant factor in their day-to-day decision-making. The casual/social drinker confronts the perhaps unanticipated demands of convenient necessity to drive after drinking and in most cases their decision is bolstered by the lessons of experience that they have done this before and not been caught. At best they have imperfect information about their BAC level, the likelihood of apprehension and even the total sanctions if arrested and convicted. It may be that harsh punishment alone is not the most effective sanction against either group of drunk driver if reducing recidivism and the incidence of DUI are the primary goals. In either case the goals and underlying assumptions of our current DUI policies should be clarified.

Are all impaired drivers equal?-Is it just about social protection?

There is no debate that the alcohol-impaired driver is a significant social problem. As a matter of law and policy the only meaningful debate is on how to formulate and implement the most effective response. In this section we examine our responses to DUI within the context of the more general problem of driving while distracted (impaired) in an effort to think clearly not just about what we do, but about the forces and assumptions behind our policy decisions.

In 2006 the Center for Disease Control reported that 13,407 people died in alcohol-impaired driving crashes, accounting for just over 30 percent of all traffic related deaths (CDC). A recent, more limited study on driving behavior found that 80 percent of crashes and 65 percent of near-crashes involved driver inattention, and that the most common form of driver inattention or distraction was the use of cell phones. Data reported by the Human Factors and Ergonomics Society estimates that cell phone distraction while driving causes over 2,500 deaths and as many as 330,000 injuries each year. Additional studies have found a substantial decline in a driver's reaction time when talking on a cell phone, whether or not hands-free. Some investigators have found that drivers talking on a cell phone have slower reaction times than drivers with blood alcohol levels exceeding 0.08. The potential future impact of this is enormous when you consider that cell phone ownership is near universal and multiple surveys estimate that over 70 percent of those with cell phones use their phone while driving.

The point of this discussion is not to quibble over which hazard is greater but to ask why we condemn the drunk driver, but seem to rationalize the behavior of the talking driver. And how does this different perspective influence our policies toward each type of driver? Perhaps our policy focus should be more expansive and DUI could be subsumed under a more general category of "impaired driving." Regardless of these future policy changes, the immediate need should be focused on making the necessary changes to allow the AWC to expand to more effectively address the problem of DUI.

Recommended Changes to Expand Participation in the AWC

Recommendation One. Reorient and/or clarify our policy goals in dealing with drunk drivers. This is a necessary first step. Is it possible to separate the issues of alcohol abuse/addiction from driving safety on a policy level? Our current policies as implemented can be seen as punishing alcohol abusers in the hopes of making them more responsible drivers and/or punishing drivers in the hopes of treating their alcohol abuse. In this sense our current policies are ineffective strategies unlikely to eliminate DUI and improve traffic safety or to reduce alcohol abuse/addiction and improve individual and community health. An essential change that would allow for more focused interventions is the development of timely and comprehensive assessment of all drivers arrested for DUI. The more comprehensive assessment would assess level of alcohol use and individual readiness for change as an indicator of amenability to mandatory treatment. Incorporating the most recent research-based information, one goal would be to distinguish those arrested for DUI who generally drink at low-risk levels from those with drinking patterns that make them at-risk for additional alcohol-related problems. Some of these factors are outlined in the National Institutes of Health 2009 publication "Rethinking Drinking: Alcohol and Your Health."

Assuming the AWC continues with limited resources of funding and staff, a primary first order goal of the assessment is to help select those offenders most likely to benefit from the AWC intervention. There are many factors to consider as a prelude to any therapeutic intervention, but the minimum should include: blood alcohol concentration (BAC) at time of arrest, self-reported drinking patterns and consumption levels, some measure of readiness for change/amenability to treatment, and history of other alcohol-related problems. Certainly the AWC already collects and uses this information to inform treatment in some manner, but the suggested process would additionally use the information to help determine who should enter the AWC. Importantly, it would separate DUI offenders who drink at low-risk levels from those who drink at levels making them at-risk of many alcohol-related problems including DUI. National survey data on alcohol use by adults indicates that about 28 percent of drinkers do so at heavy or at-risk levels. State data for Alaska suggests that this level is higher here. (See Table 4 and Figure 1.) In the context of DUI policy this component of the assessment would distinguish two large groups of offenders, those identified with low-risk drinking and those with at-risk or heavy drinking. Related research and the results of the current evaluation of the AWC indicate that the AWC treatment program is most effective for those who drink at heavy or at-risk levels. An alternative treatment program should be developed for those DUI arrestees who drink at low-risk levels that addresses their particular needs.

Table 4. Trends in Alcohol Use Among Adults, by Gender, Alaska Behavioral Risk Factor Surveillance System (BRFSS)
Figure 1. Percentage of Adults Reporting Selected Health Risks in Anchorage and Mat-Su, Fairbanks, Alaska Statewide, and the U.S., 2005

Recommendation Two. Develop alternative treatment programs and expand the number of available treatment providers. For either group of DUI offenders and AWC participants, the specific mandatory treatment must expand with resources proportionate to the problem. The AWC currently has an operating capacity of 80 participants but there is an insufficient number of treatment providers in the community for the AWC to reach this goal. This is both a policy and programmatic change. Recognizing that not all DUI offenders are the same, and therefore not all benefit from the same treatment program, requires programmatic change. Ideally, there needs to be a balance between the capacity of the court and community treatment resources. An imbalance on either side undermines both the treatment effectiveness and cost/benefit efficiency of the court. If the court lacks available space, motivated participants will be turned away. If the community lacks sufficient treatment resources, the court will operate under capacity. In either case the offender, the court, and the community are not well served. The state, local communities, and political leaders have primary responsibility to ensure adequate funding for the court and to develop policies that encourage and support the development of treatment programs within the community.

Recommendation Three. Establish a five year follow-up program with the potential for offenders to have their court record (arrest and conviction) of DUI cleared at some point, coupled with ongoing incentives to maintain desistance. This would require a change in law and a change in the AWC treatment program. Most drug treatment courts do not include alcohol-related offenses such as DUI nor do they typically concentrate on offenders whose primary drug of choice is alcohol. Historically, some drug courts operated on a pre-adjudication model such that when the participant successfully completed the treatment program, their arrest was erased and as a result there was no record of a conviction or plea. The underlying rationale for this drug court model was that by diverting the offender from the standard criminal justice sanction, the stigma of arrest and collateral negative consequences of the criminal justice process were mitigated, the offender's underlying problem could be addressed, and costs were reduced. Currently the vast majority of DTCs require the participant to enter a plea agreement and admit guilt as a condition of entry-although in many states completion of a treatment program, rather than incarceration, is the mandatory sanction for many drug offenses. Additionally, many states provide a mechanism through which the criminal record of the participant is eliminated if they successfully complete the treatment program. It seems clear that many participants in the standard DTC are motivated by these incentives. The AWC is not able to offer the benefit of a clean record and fresh start as an incentive to increase participation and support continued desistance. The AWC treatment program lasts at least 18 months, absent any false starts or relapse that may extend the program length. If the participant maintains desistance from alcohol or at a minimum is not arrested for another alcohol-related offense during the 24-36 months following completion of the program, some type of positive response seems reasonable. The response could range from expunging their court record to a resetting of sanctions to a first offense for any subsequent convictions. Some type of contact would occur every four to six months throughout the follow-up period.

Recommendation Four. Develop proactive policies of prevention to confront the problem of alcohol-impaired driving and the underlying problem of alcohol abuse and addiction. No matter how many treatment resources are provided and how effectively and efficiently they are implemented, these policies and programs are primarily reactive and as such can never solve the problem. Ultimately, when the community of Anchorage and the State of Alaska commit to a comprehensive policy of prevention infused across the spectrum of private and public institutions, there will be less demand for treatment programs like the AWC and the sanctions of the criminal justice system. The greater need for the coercive powers of the criminal justice system reflects a degree of failure throughout the basic institutions of a community and society.


Problem-solving courts created to process drug offenses and enhance the administrative efficiency of criminal courts have become an essential mechanism to reduce recidivism rates and address the needs of drug offenders. Drug treatment courts have operated for nearly 20 years and many evaluations of their operation have been completed on the local, state and national level. In this span of time researchers, criminal justice professionals and treatment providers have learned much to increase the effectiveness and efficiency of drug treatment courts, but significant issues remain unresolved. Perhaps the most significant issue concerns the ability of the drug court model, employing the tenets of therapeutic jurisprudence, to include a larger percentage of drug-involved offenders. As mentioned in the previous discussion, most DTCs including the AWC enroll only a small percentage of eligible offenders. In many traditional drug courts, otherwise eligible offenders are intentionally excluded because of specific offense histories, most often violent offenses. The AWC has eligibility guidelines, but this is not the primary factor hindering enrollment. For various reasons, and most remain unclear, only a small percentage of eligible offenders decide to participate. Capitalizing on the research evidence documenting that coerced treatment can be as effective as voluntary treatment, the AWC could easily expand through a screening process weighted more on amenability assessment than voluntary self-selection.

The AWC has achieved a degree of success over the past decade and is a well established therapeutic court within Anchorage and Alaska. The continued success of the AWC is uncertain and will be determined to some degree by how well it is able to address the issues presented in this essay.

Ronald S. Everett is an Associate Professor with the Justice Center, and assisted with the research for the Urban Institute Justice Policy Center report, "Impact and Cost-Benefit Analysis of the Anchorage Wellness Court" published in 2008.

The Anchorage Wellness Court evaluation was a collaboration between the University of Alaska Anchorage Justice Center, the Urban Institute, and the Alaska Judicial Council. This project was funded by Grant No. 2003-DD-BX-1015 awarded by the National Institute of Justice, Office of Justice Programs, U.S. Department of Justice. Points of view in this document are those of the authors and do not necessarily represent the official position or policies of the U.S. Department of Justice.