Milwaukee County is the largest of Wisconsin’s counties. Its county seat is the city of Milwaukee. The county has a population of 957,735 (2015 U.S. Census Bureau). Its racial makeup is 65% white, 27% black or African American, and 14% Hispanic or Latino of any race (2015 U.S. Census Bureau). The median household income is $43,873, and 22.0% of the population live at or below the poverty line (2015 U.S. Census Bureau).
In calendar year 2015, 5,630 felony and 4,485 misdemeanor complaints were filed, in addition to 2,566 criminal traffic cases. The circuit court is composed of 47 nonpartisan circuit court judges elected by voters, 14 of whom handle felony cases and 7 of whom handle misdemeanors (Wisconsin Court System, 2016). The elected district attorney has approximately 125 assistants, while approximately 60 public defense attorneys handle adult criminal matters and 25 public defense attorneys handle juvenile, Children in Need of Protection or Services (CHIPS), Termination of Parental Rights (TPR), and related cases under the auspices of the Wisconsin State Public Defender’s Office. Jail bookings in calendar years 2008, 2011, and 2014 were 42,179, 40,043, and 33,515, respectively.
|Profile of Milwaukee County’s Justice System||
|Jail and House of Correction Rated Capacity||
|Jail Average Daily Population||
|Felony Cases Filed||
|Misdemeanor Cases Filed||
|Adult Probation Population||
Several years prior to their selection as an EBDM pilot site, Milwaukee County formed a Community Justice Council (CJC) to ensure that efforts to reduce crime, support victims, and hold offenders accountable were coordinated and collaborative. The council was formed by county board resolution and is comprised of 27 members representing state and local government, criminal justice, and social services. The council’s focus on strategic planning and research to identify strategies to improve the justice system made the work of implementing the EBDM Framework a natural fit for the county.
MILWAUKEE COUNTY’S EBDM VISION, MISSION, AND TEAM
The Milwaukee County EBDM Policy Team was formed in 2010 as a subset of the executive committee of the Milwaukee County Community Justice Council. The team includes the following elected officials and stakeholders:
- the county executive;
- the sheriff;
- a county board member;
- the district attorney;
- the city mayor;
- the chief judge of Milwaukee County Circuit Court, First Judicial District;
- the city’s chief of police;
- the State of Wisconsin first assistant public defender;
- the executive director of the Benedict Center;
- representatives from the Department of Corrections and the State Office of Justice Assistance;
- the United States Marshal of the Eastern District of Wisconsin (ex officio);
- the court’s pretrial services coordinator;
- an inspector from the sheriff’s office; and
- the presiding judge of the felony division of the Milwaukee County Circuit Court.
The Milwaukee County EBDM Policy Team’s vision and mission statement reflects its commitment to public safety achieved through collaboration and the use of research to support decision making: “By applying what the evidence tells us about what actually works in protecting the community and holding offenders accountable, Milwaukee County’s criminal justice system will make the smartest possible use of its limited resources, continuously improving its performance against quantifiable goals and reinvesting the savings in programs that reduce crime in the first place.”
In Phases II and III of EBDM, the policy team was staffed by the coordinator of the Community Justice Council, a deputy district attorney, and a public defender. This was a strategic decision designed to ensure that the team’s EBDM work was built upon a foundation of strong collaboration between prosecution and defense (among other parties).
MILWAUKEE COUNTY’S CHANGE TARGETS
The early activities of the Milwaukee County EBDM Policy Team focused on establishing a high functioning collaborative team and following the EBDM Phase II planning roadmap. The roadmap assisted them with assessing their policies and practices at each of the EBDM decision points, determining the degree to which research evidence guided their decisions, and identifying strengths, challenges, and targets for future policy and practice change. Three of the areas that the team deemed to offer the greatest opportunity for improving risk and harm reduction outcomes were:
- developing a pretrial management process that would ensure that release/detention decisions of pretrial defendants were informed by current research;
- revising and expanding diversion and deferred prosecution (or “early intervention”) options to avoid traditional case processing, where appropriate, and to maximize the opportunity to support and encourage prosocial attitudes and behaviors among those who become involved in the justice system while minimizing the potential negative consequences that accrue to an individual involved in the system; and
- developing and piloting a new approach to sentencing and probation supervision that allows probationers the opportunity to earn early termination from supervision by completing risk reducing interventions.
This case study offers a review of Milwaukee County’s Phase II/III efforts as they relate to these three change targets. For more information on the full range of change targets developed, see Milwaukee County’s Phase III implementation plan.
WHAT THEY DID: EXPANDED PRETRIAL SCREENING; DEVELOPED A PRAXIS TO GUIDE PRETRIAL RELEASE DECISION MAKING
“One of our main priorities is to apply evidence-based decision making to pretrial release and detention determinations. While public safety is always our primary concern, we want to ensure that we are being good stewards of public funds and making the best use of limited community resources. We know that risk is inherent in all pretrial release decisions, so we want to base our determinations more on the risk posed by a particular defendant rather than the charge he is facing. We hope to minimize the disparity in release decisions for similarly situated defendants.”
—Jeffrey Kremers, Chief Judge, Milwaukee County
Making pretrial release and detention decisions without actuarial assessment information can have deleterious effects on both defendants and the general public. Releasing extremely high risk defendants without assessment or matched supervision can be a threat to public safety. At the same time, recent research has demonstrated that detaining low and moderate risk defendants in jail for even short periods of time (i.e., 2–3 days) can increase their risk for misconduct both short- and long-term. During Phase III of EBDM, Milwaukee officials took several important steps to ensure that pretrial release and detention decisions were informed by information about defendants and contextualized by these considerations.
First, in 2011 they validated and reconstructed their pretrial risk assessment tool (the Milwaukee County Pretrial Risk Assessment Instrument [MCPRAI]), which had been in use for several years. The resulting MCPRAI-Revised uses six predictors of pretrial outcomes, including cases filed, prior failures to appear, arrest while on bond, employment/caregiver status, residence, and substance abuse to determine an individual’s risk for pretrial misconduct (i.e., failure to appear in court and new criminal activity). Second, they instituted a process for conducting pretrial screening of all detained bailable defendants (“universal screening”), comprising approximately 98% of Milwaukee County arrests. Third, they developed a decision making tool (the Milwaukee County Pretrial Praxis, hereafter “the Praxis”) to guide both release and supervision decisions. The Praxis is a decision making framework used to prescribe bond type, supervision level, and supervised conditions during the pretrial stage of the case. The Praxis ensures that bail decisions are based on the defendant’s risk of committing a new crime and failure to appear, as well as his or her individual criminogenic needs. Fourth, to ensure broad support for changes, systemwide training was conducted over many months. (The initial training event included 200 representatives from the court, district attorney’s office, and public defender’s office, among others; a subsequent event included 325 representatives from agencies throughout the justice system community; and smaller meetings were convened within the court and the district attorney and public defender’s offices. These and other events served as venues to explore relevant research, discuss the changes underway in the county, and provide opportunities for stakeholder discussion and feedback.) Fifth, to confirm that the MCPRAI-R was predictive for the jurisdiction’s population, a validation study was conducted in 2012 using data from approximately 3,500 individuals who had been booked into the Milwaukee County Jail on criminal charges. The results of the study indicated that the instrument was, in fact, accurately predicting pretrial failure.
In addition, in May 2012, the EBDM policy team conducted a survey of stakeholders (i.e., court officials, prosecutors, public defenders, and pretrial) to gauge their level of satisfaction with the use of the pretrial risk assessment instrument, the Praxis, and accompanying pretrial processes. Survey results were used to inform further training and to refine pretrial practices. The results and key findings of both the validation study and stakeholder survey were presented to all stakeholder groups in a day-long session in May 2013.
Highlights from the 2012 validation study and stakeholder survey include the following:
- 80% of survey respondents indicated that implementation of the MCPRAI-R provides stakeholders with better and more consistent information to make decisions on pretrial release. 67%–80% of respondents indicated they believe the MCPRAI-R accurately predicts the risk for pretrial failure and that the Praxis leads to sound decisions about release.
- Milwaukee County had a 75% pretrial release rate.
- Judicial officers adhered to the Praxis-recommended bail type and supervision in 78% of cases.
- 87% of pretrial felony and 84% of pretrial misdemeanor defendants had no new criminal activity during the pretrial period.
- 84% of pretrial felony and 62% of pretrial misdemeanor defendants appeared for all scheduled court dates.
- The MCPRAI-R was accurately predicting pretrial failure.
- There was a relatively high failure to appear rate for misdemeanor defendants who fell into Risk Level IV. (In order to improve outcomes for this group, recommendations were made and subsequently implemented to modify recommended conditions of release.)
More recent outcome data is lacking due to significant structural issues with Milwaukee’s very antiquated pretrial information system. To remedy the ongoing data challenges, a new pretrial data and case management system has been constructed. The system will include much more robust data-reporting capabilities and is expected to be implemented in early 2017.
Milwaukee County’s “early intervention (EI)” strategy is a set of diversionary options designed to follow key research principles—primarily the risk and need principles—and reduce the time, cost, and collateral consequences associated with traditional processing in appropriate cases.
The goal of these EI programs is to reduce the long-term recidivism risk of individuals involved in the criminal justice system while at the same time ensuring public safety and the efficient allocation of limited criminal justice resources.
The programs seek to maximize the opportunity to support and encourage prosocial attitudes and behaviors among those who become involved in the justice system while minimizing the potential negative consequences that may accrue to an individual involved in the system, such as social stigma, exposure to higher risk offenders, and loss of prosocial supports (family, employment, educational activities, etc.).
Dating back to March 2007, the Milwaukee County district attorney’s office had worked in collaboration with the public defender’s office to identify suitable defendants for diversion and deferred prosecution; approximately 117 defendants per year were provided the opportunity to avoid traditional case processing, criminal charges, and prosecution. However, as stated in the words of one Milwaukee official during the course of EBDM discussions, “It was a very non-uniform process…A lot of these decisions were just gut-based.” Identification of Diversion and Deferred Prosecution Agreement (DPA) cases was dependent upon the assigned assistant district attorney and/or defense attorney, and formal eligibility criteria and conditions associated with participation in each option had not been formulated. The system was not informed by risk in terms of selection criteria or by social science research in terms of program expectations. As a result, a key area of focus for Milwaukee’s EBDM work included a redesign and expansion of diversion and deferred prosecution and the development of an “early intervention (EI)” strategy.
Under the EBDM initiative—and then continued with funding provided by a U.S Department of Justice, Bureau of Justice Assistance, Justice Reinvestment Initiative-Local (JRI) grant—Milwaukee officials developed a conceptual framework for the early intervention strategy and a comprehensive document articulating the policies, procedures, and protocols to support the strategy. The EI strategy includes:
- integrating a risk/needs assessment process for early intervention options into the universal screening approach applied to pretrial defendants;
- redefining the eligibility requirements for both the diversion and deferred prosecution tracks;
- tailoring the programmatic requirements of each track to the risk level of the target population, with a “minimum intervention” approach for the low risk diversion participants and a risk reduction approach for the moderate risk deferred prosecution participants; and
- creating a “Central Liaison Unit (CLU)” to provide case management and, for the moderate risk population, risk reduction interventions.
An important modification to their assessment process included adding the Level of Service Inventory-Revised Screening Version (LSI-R:SV) and, in some cases, the Level of Service Inventory-Revised, to the universal screening protocol. While these tools are not relevant to the pretrial release/detain decision, they do serve to identify potential eligibility for diversion and deferred prosecution eligibility. Given that diversion was established as a pre-charge option for low risk defendants, obtaining this information prior to charges being filed is critical.
Diversion and deferred prosecution eligibility and program requirements were established as follows:
- To be eligible for the diversion program, a defendant must be screened as low risk (an LSI-R: SV score of 0–2) and have pending charges for non-exclusionary crimes. Individuals who enter the diversion program must meet specific program expectations and remain crime-free for the term of the diversion (typically 6 months). Since participating defendants are assessed as low risk, conditions are focused on victim restoration (e.g., victim–offender mediation, letters of apology, restitution) and/or accountability (e.g., community service, educational programs or classes, referrals for school/job training) rather than risk reduction. The program is designed to encourage defendants to continue or pursue prosocial activities. Participants in the diversion program are required to have legal counsel and complete and sign a written Diversion Agreement. If the defendant satisfactorily completes the terms of the agreement, charges are not filed by the district attorney. If the defendant does not satisfactorily complete the terms of the agreement, the district attorney issues the criminal charges and proceeds with formal prosecution.
- Individuals who are assessed as moderate risk (i.e., those who have an LSI-R:SV score greater than 2 and are then assessed as moderate on the LSI-R) and are charged with non-exclusionary crimes are eligible to enter into a Deferred Prosecution Agreement (DPA). The conditions of a DPA must address risk reduction (e.g., cognitive behavioral therapy [CBT], substance abuse/mental health treatment, anger management with a CBT component), although they may also include accountability and/or restoration strategies (e.g., drug testing, restitution, electronic monitoring/GPS). Conditions are focused on addressing the defendant’s top 1–3 criminogenic needs, and over-conditioning is avoided. DPA participants are assessed and routinely monitored by case managers; progress/compliance reports are required initially at three months and subsequently at six-month intervals. Prosocial behaviors are intentionally encouraged and recognized. If there is noncompliance with program conditions, a written plan is put in place to address the noncompliance. DPA duration is based on risk and need, but is generally between 6 and 18 months. DPA participants are considered successful if they complete the requirements as agreed and remain crime-free during the period of monitoring under the DPA. If the individual satisfactorily completes the program, the formal charges are either dismissed or reduced. If the individual does not satisfactorily complete the deferred prosecution program, the individual is sentenced on the charge(s) as pled.
Through careful protocol development, Milwaukee County successfully re-engineered its approach to diversion and deferred prosecution in an effort to more closely align its services with evidence-based practices: decreasing the level of intervention with the low risk population while delivering, as early as possible, risk reduction services to the moderate risk. From 2007, when the first diversion and DPA programs were developed, to 2012, slightly more than 70% of participants successfully completed Diversion/DPA Agreements. An evaluation of data from March 2014 to June 2016 (the most current data available) demonstrates that the EI strategy redesign both increased the number of defendants participating in these programs and improved their outcomes. Specifically, more than 1,100 individuals were served through these early intervention programs (808 through diversion and 331 through deferred prosecution) during the 15-month study period, the vast majority of whom completed their agreements successfully: 81% of diversion cases satisfied their agreement with no charges filed, and 77% of DPA cases had charges dismissed or reduced. Less than 3% of diversion participants and less than 8% of DPA participants had a new arrest during the agreement period. Finally, jail beds for this population were significantly reduced, with 24,891 jail days saved as a result of the diversion program and 16,835 saved as a result of the DPA program (calculations based upon the recommended sentence by the state at the time the agreement is entered). For more information, see the report Milwaukee County, Wisconsin, Early Intervention Strategy: A Case Study in Evidence-Based Diversionary Practices.
Dosage Probation Pilot
The project seeks to:
- lower recidivism among the target population;
- reduce the average length of supervision for those who successfully complete supervision; and
- achieve system alignment by aligning sentences with risk/needs assessment information, supervision practices with risk reduction-focused sentencing, community-based treatment options with criminogenic needs, and incentives with the achievement of risk reduction goals.
- Reduce new charge arrests, technical violations, and revocations
- Shorten the time that individuals are on probation
- Cut costs of probation by 15%
- Reduce probation recidivism by 15%
For more information, visit: http://milwaukee.gov/cjc/CJCCommunityInitiatives/Dosage-Probation.htm
During their Phase II work, in concert with their EBDM TA provider, the Milwaukee County EBDM Policy Team conceived of a new model of sentencing and probation called “Dosage Probation.” The model is designed to more closely align probation sentencing practices with risk reduction research; its ultimate goal is to improve offender outcomes (i.e., recidivism reduction). The key elements of the model include incentivizing offenders’ engagement in risk reducing activities, ensuring offenders receive interventions and services that have been demonstrated effective in reducing recidivism, and providing the opportunity for early termination from supervision when risk reduction goals (i.e., dosage targets, based upon risk level) have been met.
One of the first planning activities included developing a logic model to identify key activities and anticipated outputs and outcomes for the project. The team also came to agreement on the dosage probation process, including eligibility criteria and how offenders are assigned to the dosage probation unit. Other key steps included developing intervention tools to support dosage probation (e.g., behavior response guidelines, staff behavioral checklist, offender behavioral checklist) and developing and conducting training and coaching activities for the dosage probation officers (i.e., “the four core competencies,” Motivational Interviewing, use of risk reduction tools and worksheets).
Following the initial implementation of the dosage model, it was expanded to a second probation unit in Milwaukee. In addition, NIC has funded a pilot project to test the model in two jurisdictions outside of Wisconsin; a research study to demonstrate the efficacy of the model may be funded in the future. In Phase VI of their EBDM work, the Wisconsin State EBDM Policy Team is considering the potential of expansion of this model to other probation units in the state.
In 2015, Milwaukee County decided to take another step to improve its pretrial justice system by seeking to become an implementation site for the Laura and John Arnold Foundation’s Public Safety Assessment. Milwaukee County was ultimately selected, and after approximately one year of technical assistance, full implementation occurred in June 2016. This work also included development of a new pretrial decision-making framework. The decision to change from the MCPRAI-R to the PSA was based on the robust research and data behind the development of the PSA. In addition, the PSA does not use risk factors such as employment, residence, and substance abuse that could lead to disparities. This change marks an ongoing commitment by Milwaukee County justice system stakeholders to continuous system improvement by implementing the most current, evidence-based risk assessment instrument available.
In 2016, Milwaukee County was selected to receive $2 million from the John D. and Catherine T. MacArthur Foundation as a participant in the Safety + Justice Challenge. Their efforts build on previous EBDM and JRI reform efforts and aim to reduce the length of jail stays for low level nonviolent misdemeanants, divert individuals with mental health and substance abuse issues, and provide more trauma-informed response options to professionals and community members.
As Wisconsin advances its EBDM efforts through Phase VI, Milwaukee County representatives continue their contribution to the Wisconsin State EBDM Policy Team’s efforts and, at the same time, will continue to meet to advance local strategies to improve performance and reduce harm in their local community.