The criminal justice system’s experience with COVID-19 quickly exacerbated the problems of overcrowded jails. As case processing slowed and with corrections officials reluctant to accept new inmates into prisons, local jails became even more crowded. Innovative ways to monitor inmates in the community helped to reduce jail numbers and can continue to do so after the pandemic. Among other things, pretrial release programs ameliorated COVID-related overcrowding by reducing the number of inmates sitting in jails awaiting trial simply because they could not post a low bond amount. That is, counties that had well-developed pretrial programs were able to rely on them to a greater degree during the pandemic. Although the COVID-19 virus did not cause these programs to come into existence, it has accelerated their use in counties that had them and demonstrated their worth to the administration of justice.

Pretrial programs have the potential to reduce jail populations and, more important, help people begin a rehabilitation process before their case is resolved. This article explores Wisconsin’s use of pretrial programs, the benefits they can provide to the public and the justice system, and methods by which they can be brought into existence statewide.

National Institute of Corrections Project

In September and October 2020, the Wisconsin Court System’s Office of Court Operations¹ conducted a statewide survey to determine which counties engaged in pretrial services, how those programs are viewed, and what training might be useful to counties regarding such services.² In addition to providing answers to questions about the prevalence of such services, the survey also identifed potential grant opportunities for the future.³ Responses were received from 46 of Wisconsin’s 72 counties – a 64 percent response rate. Of the responding counties without pretrial services, a majority indicated there were no plans to create a pretrial services agency.

On a national level, there has been an effort underway for quite some time to move toward pretrial release programming rather than rely on cash bail.⁴ A congressional report from as early as the mid-1960s noted the following:

“There is no doubt, however, that each year thousands of citizens accused of crimes are confined before their innocence or guilt has been determined by a court of law, not because there is any substantial doubt that they will appear for trial but merely because they cannot afford money bail. There is little disagreement that this system is indefensible.”⁵

In 2008, the National Institute of Corrections (NIC) launched a program called the Evidence-Based Decision Making Initiative (hereinafter the EBDM Initiative).⁶ The NIC is a federal agency within the U.S. Department of Justice whose task is to provide training, technical assistance, information services, and policy and program development assistance to federal, state, and local justice system partners. The stated purpose of the EBDM Initiative was to create and implement a framework designed to improve outcomes in the justice system through collaborative partnerships and shared vision of desired outcomes. The EBDM Initiative was based on more than two decades of research on factors that contribute to criminal reoffending and methods that justice systems can use to interrupt the cycle of crime.⁷ Pretrial release programming was a part of this initiative.

In August 2010, seven jurisdictions from throughout the United States were chosen to pilot the EBDM Initiative. Milwaukee County and Eau Claire County were among those sites.⁸ For the next two years, support was provided to EBDM pilot sites as they implemented their plans for meeting locally identified outcomes. Strategies included implementing and validating local pretrial risk-assessment tools, implementing universal screening of pretrial defendants, adding or redesigning diversionary policies, and aligning interventions for individuals based on level of risk and needs.⁹ Thus, a great deal of focus was placed on pretrial issues.

A case study of the original seven national sites was released in 2017. This case study reviewed the data for some, but not all, of the targeted improvements for the various pilot sites throughout the country.¹⁰ Although collecting data on pilot projects can be complicated, this case study provides some useful information.

For Eau Claire County, the case study specifically reviewed data from the implementation of a risk-assessment screening tool and data from the county’s diversion program. With respect to the implementation of a risk-assessment screening tool, Eau Claire County had begun using a time-of-arrest tool called the Proxy.¹¹ In gauging success, they sought ongoing feedback about the implementation from law enforcement agencies. A county-conducted study comparing the Proxy results and the COMPAS¹² results found a very high correlation between a low-risk score on the Proxy and a low-risk score determined by the COMPAS, confirming use of the Proxy as a pretrial triage tool. Regarding the Eau Claire County diversion program, the case study found that recidivism rates were substantially reduced for individuals participating in the diversion program in comparison to individuals with similar charges prosecuted through the district attorney’s office. Thus, initial results for these two aspects of Eau Claire County’s “pretrial” change targets seemed to be positive.

For Milwaukee County, the case study reviewed three change targets: 1) development of a pretrial management process to ensure that release and detention decisions were informed by current research, 2) revising and expanding the diversion and deferred prosecution options, and 3) developing and piloting a new approach to sentencing and probation supervision. As part of the data-collection process, Milwaukee County conducted a stakeholder survey regarding pretrial goals. Eighty percent of the survey respondents indicated that the implementation of the pretrial risk-assessment tool (the Milwaukee County Pretrial Risk Assessment Instrument [MCPRAI]) provided stakeholders with better and more consistent information to make decisions on pretrial release. Sixty-seven to eighty percent of the same respondents indicated that the MCPRAI, and the decision-making tool that is used to guide release and supervision (the Praxis), accurately predicted the risk for pretrial failure.

The case study also revealed that the redesign of the diversion and deferred prosecution processes increased the number of defendants participating in these programs and improved their outcomes. Although not pretrial focused, Milwaukee County did develop a new model of sentencing and probation, which since has been funded as a pilot project to test in two jurisdictions outside Wisconsin. Thus, Milwaukee’s “pretrial” change targets also seemed to demonstrate success.

Given the positive results received from the initial seven sites nationwide, in 2015 an additional 21 jurisdictions were added nationwide. The following Wisconsin counties were brought into the project: Chippewa, Outagamie, Rock, Marathon, La Crosse, and Waukesha.¹³ Because Milwaukee and Eau Claire counties had been in the program for a longer time, the goal for the additional counties was to bring them along in the path of the two earlier counties.

A “Non-EBDM Initiative” Case Study

In 2017, the Brown County Criminal Justice Coordinating Board (CJCB) began to focus on the issue of overcrowding in the Brown County jail. At that time, the CJCB focused on a component of the jail population over which a measure of control could be reasonably obtained – individuals who were in jail while awaiting sentencing.¹⁴ Aware that Outagamie County was participating in the EBDM Initiative, in the fall of 2017 the Brown County CJCB approached Outagamie County in an effort to gain insight into the latter county’s level of success with this population as a result of its efforts with the EBDM Initiative.

By the time Brown County reached out, however, the assistance from the national pilot project was ending. A limitation of the pilot project concept is that typically, if the pilot project sites prove to be successful based on grant funding or heightened assistance from federal or state governments, the pilot cannot be replicated in non-pilot sites because of lack of resources. Brown County faced this obstacle. Thus, no assistance from the NIC was available to Brown County, but a neighboring county was ready and willing to share its experience.

The Brown County CJCB members needed not only resources but also assistance in gaining buy-in from stakeholders. The cost of building new jails, or at least the cost of housing inmates, is always of interest to local politicians. It is a difficult sell to put money on the table for new and better accommodations for people accused or convicted of breaking the law. Thus, focusing on this issue made sense to all stakeholders.

The Brown County CJCB members sought, among other things, to create a pretrial criminal justice services division, including staff, either as a stand-alone department or as a division of the Health and Human Services Department; to implement a screening tool that would assist decision-makers in pretrial release decisions; to strengthen the pretrial diversion program; and to bring the privately contracted day-report center in house to streamline communication with the court, the prosecutor, and the defense attorneys.¹⁵

Looking at pretrial options seems like sound constitutional and statutory policy for the justice system, defendants, crime victims, and the public at large.

In 2019, with the support of the Brown County Board of Supervisors, the Criminal Justice Services (CJS) Division of the Brown County Health and Human Services Department was created with sufficient funding for staff. This funding piece came largely from the savings on the elimination of the outside contract for day-report services, which were brought in house to the CJS Division. Thus, these two change targets were very interconnected and the accomplishment of one meant the attainment of the other.

A CJCB subcommittee concurrently started to review various risk-assessment tools.¹⁶ Around this time, the Laura and John Arnold Foundation was releasing its newly developed assessment tool, called the public safety assessment (PSA). This tool is based on data from 750,000 cases from 300 jurisdictions throughout the United States.¹⁷ The CJS selected this tool and immediately began implementing this risk-assessment tool for pretrial defendants.

Among other things, the CJS provides PSA reports to the court for use in all misdemeanor and felony cases coming before the court for bail decisions. Between May 1, 2019 and May 1, 2020, the CJS generated 3,122 PSA reports and provided them to the court for assistance in bail decisions. In addition, the CJS provides status reports for anyone being monitored in the community on bail. The Brown County diversion program also saw a significant uptick in usage – meaning many first-time offenders were screened out for community service or treatment in exchange for no prosecution. Restorative justice programming can also play a role in diversion programming, thereby aiding crime victims and society as a whole.

During the COVID-19 pandemic, Brown County has been able to ameliorate jail crowding by relying heavily on the CJS day-report services to monitor defendants released pretrial. As stated above, COVID-19 did not create pretrial programming, but the pandemic has afforded an opportunity to demonstrate the programming’s worth.

The Case for Pretrial Services

A simple demonstration that a given county, such as Brown, Milwaukee, or Eau Claire, is using pretrial services does not mean that they work nor does it mean that other counties should be obligated to use them. However, the Wisconsin Constitution actually addresses this issue:

“All persons, before conviction, shall be eligible for release under reasonable conditions designed to assure their appearance in court, protect members of the community from serious bodily harm or prevent the intimidation of witnesses. Monetary conditions of release may be imposed at or after the initial appearance only upon a finding that there is a reasonable basis to believe that the conditions are necessary to assure appearance in court.”¹⁸

Wisconsin statutory provisions are similar. Thus, looking at pretrial options seems like sound constitutional and statutory policy for the justice system, defendants, crime victims, and the public at large. Of the 46 counties that responded to the pretrial study conducted by the Office of Court Operations, 19 offer pretrial services and 27 do not.¹⁹ Although a complete discussion of all the counties that focus on pretrial services and the data regarding results is beyond the scope of this article, the focus of this discussion is to demonstrate that it is not beyond the scope of any Wisconsin county, even the 27 counties that responded but do not offer those services and the 26 that did not respond at all, to firm up their pretrial procedures and take advantage of some of the evidence-based processes that have already been piloted in Wisconsin. Smaller and more rural counties can benefit from these ideas and services as well as larger urban counties.

Although the types of services offered may vary in such programs, several characteristics seem to surround success in this realm: risk assessment, some form of monitoring program for those released into the community pending trial, and some central agency or employee to run the pretrial program and take responsibility for its success.

Assessing Risk of Not Showing Up for Court or Reoffending

A vital piece of a successful pretrial release program is an understanding of the risks involved with releasing the defendant back into the community pending trial. The risks are twofold: not showing up for court and reoffending. Further, making release decisions using an evidence-based assessment tool with supporting data helps with buy-in from stakeholders and the public. In Brown County, there was some resistance to the use of a risk-assessment tool because of a stated concern that the community deserves to have such decisions made by an elected judge (or appointed court commissioner) rather than by a tool that uses data about criminogenic factors. That is, when a release decision goes bad, there should be a person who can answer for the decision rather than data in some distantly conducted study.

There are never guarantees with such calculations, but if one starts with an assumption that people who do not need to be in the jail should be in the community and might be harmed if they are in the jail unnecessarily, then the question is more about the best way to become smarter at setting bail rather than simply the willingness of a magistrate to “own” a poor bail decision. Having additional information, in the form of a risk assessment, seems to be an improvement in the bail system.

There is ongoing debate about the extent to which such tools should be relied on given variables such as geographic location, race, and economic circumstances. Nevertheless, several risk-assessment tools have been designed and are used throughout Wisconsin to facilitate smarter bail decision-making. While a full discussion of risk-assessment tools is beyond the scope of this article, a limited discussion is warranted.

The survey report for the Office of Court Operations cites several risk-assessment tools currently in use in Wisconsin. Ten of the responding Wisconsin counties reported they use the PSA. As referenced above, the PSA was developed by the Arnold Foundation, which used a dataset collected from 300 jurisdictions across the country consisting of 1.5 million cases of which only about 750,000 were used in creating the risk-assessment tool.²⁰

This assessment relies on nine data points or elements, which are weighted differently: age at current arrest, current violent offense, pending charge at time of arrest, prior misdemeanor conviction, prior felony conviction, prior violent conviction, prior failure to appear in past two years, prior failure to appear older than two years, and prior sentence to incarceration.²¹ While the number of prior convictions are weighted heavier than the other criteria, there is plenty of evidence suggesting that African-Americans are more often wrongfully convicted than whites, which means data can be skewed.²² Other data points also have the potential of being skewed. However, judges and justice system employees need to be aware that these assessments are tools, not magic bullets.

A vital piece of a successful pretrial release program is an understanding of the risks involved with releasing the defendant back into the community pending trial.

Ten of the responding Wisconsin counties reported they use the COMPAS PRRS-II. This tool was developed using data from a study of 2,831 felony defendants in Kent County, Mich.²³ The information gathered included various socioeconomic factors such as history of drug abuse, employment status, and length of time in the current community or neighborhood. Of course, a very narrow survey of cases from one county in Michigan may not be predictive of defendant behavior in Wisconsin.

Two of the responding Wisconsin counties reported they use the revised version of the Virginia Pretrial Risk Assessment Instrument (VPRAI) or the Revised VPRAI. The VPRAI was created by the Virginia Department of Criminal Justice Services.²⁴ The data for the Revised VPRAI was collected from 14,383 cases across seven different communities in Virginia. Those communities and cases had varied characteristics based on community type, population, sex, race, and socioeconomic status. Concern could arise regarding use of a Virginia tool in Wisconsin.

Two of the responding Wisconsin counties use the Milwaukee County Pretrial Risk Assessment Instrument-Revised (MCPRAI-R). This tool uses six predictors of pretrial outcomes: cases filed, prior failures to appear, arrest while on bond, employment or caregiver status, residence, and substance abuse. The validation of this tool involved 3,500 individuals who had been booked into the Milwaukee County jail on criminal charges.²⁵ Milwaukee County stopped using the MCPRAI-R when the Arnold Foundation’s PSA became available. The switch was based on the level of research that went into the PSA and data supporting the PSA.²⁶

One of the responding Wisconsin counties reported using the Level of Service Inventory-Revised (LSI-R), although the LSI was not developed as a pretrial risk-assessment tool. Each of these tools uses an algorithm to help predict, but not guarantee, behavior in a given case.

Thus, there is a wealth of experience in Wisconsin with the process of assessing risk and using those assessments for bail decisions. One size does not fit all counties and local validation might be desired, but variety in choosing a risk-assessment method helps counties that are looking to improve.

Day-Report Services Can Monitor Compliance with Bail Conditions

Day-report services are a vital piece of monitoring compliance with bail conditions. For example, recognition that a given defendant out on bail is actively using illegal drugs assists in assessing risk to the community as well as the life of the defendant. Information that a defendant has stopped reporting to a day-report center makes possible earlier responses than does waiting to hear about an absconder at the next hearing, which could be several weeks later. Day-report services can be contracted from an agency in the community or managed in house by the county. Elements of the services can include regular drug testing, regular check-ins to ensure defendants have not left the area, referrals for treatment, and management of electronic monitoring such as the SCRAM.

Benefit by Combining Pretrial Agency Services

The November 2020 study found that of the reporting counties, 12 were planning no action on pretrial services because of lack of funding. This is certainly a hard reality. However, many of the services that already exist in some counties would fall under pretrial services, and combining some of those services under one roof might provide some economies of scale. Furthermore, discussions of this issue by the Brown County CJCB focused on the need to have someone that can answer for the performance of the pretrial processes that have been set up. This is important insofar as any such services will need data collection and distribution by someone responsible for success.

Conclusion

In the comment portion of the 2020 Wisconsin Pretrial Survey, one county noted that “[s]ince beginning pretrial, we have heard nothing but good things from all stakeholders regarding pretrial services. We have been able to assist people in the front end with referral for services etc. Unfortunately, we will need continued funding to support this into the future.”

Success and stakeholder satisfaction seem to be following in the footsteps of pretrial programming in Wisconsin. These successes are not just for urban counties, but all counties. Pretrial programming serves all aspects of justice and reduces reliance on cash bail to ensure conditions of release. Although there are costs, pilot results are demonstrating the benefits to communities statewide.

» Cite this article: 94 Wis. Law. 38-44 (May 2021).

Endnotes

¹ Demonstrating a commitment to pretrial programming, in December 2019 the Wisconsin Office of Court Operations hired its first pretrial program manager, Rhonda Frank-Loron, to work specifically on pretrial programming in the courts of this state.

² Wis. Ct. Sys., Office of Ct. Ops., 2020 Wisconsin Pretrial Survey, Summary of Results, at 1 (Nov. 2020).

³ Id.

⁴ The State Bar of Wisconsin has also taken the position that efforts must be taken to move away from cash bail in favor of validated risk-assessment tools for pretrial detention decisions. State Bar of Wis., Bail Reform, www.wisbar.org/aboutus/governmentrelations/Pages/Bail-Reform.aspx (last visited April 8, 2021).

⁵ Charles Doyle, Bail: An Overview of Federal Criminal Law 3 (July 31, 2017) (citing S. Rept. 89-750 (1965) and H. Rept. 89-1541 (1966)).

⁶ Nat’l Inst. Corr., The Evidence-Based Decision Making Initiative: An Overview for Law Enforcement, at 1 (June 2017).

⁷ Nat’l Inst. Corr., EBDM Case Studies: Highlights From the Original Seven Pilot Sites, 4 (June 2017).

⁸ Id.

⁹ Id.

¹⁰ See generally id.

¹¹ Id. at 14.

¹² The COMPAS is a risk-assessment tool that carries with it some built-in credibility because it is used throughout Wisconsin by the Wisconsin Department of Corrections for assessing needs of defendants in the criminal justice system.

¹³ EBDM Case Studies, supra note 7, at 52.

¹⁴ Brown Cnty. CJCB Minutes, May 11, 2017, and July 13, 2017.

¹⁵ Brown Cnty. CJCB Minutes, Jan. 11, 2018, and Mar. 8, 2018.

¹⁶ Brown Cnty. CJCB Minutes, May 11, 2017.

¹⁷ Risk Assessment Factsheet, Public Safety Assessment (May 10, 2019), https://law.stanford.edu/pretrial-risk-assessment-tools-factsheet-project/.

¹⁸ Wis. Const. art. I, § 2.

¹⁹ 2020 Wisconsin Pretrial Survey, supra note 2, at 2.

²⁰ Risk Assessment Factsheet, supra note 17.

²¹ Kate Patrick, Arnold Foundation to Roll Out Pretrial Risk Assessment Tool Nationwide, Inside Sources (Sept. 3, 2018), https://insidesources.com/Arnold-foundation-to-roll-out-pretrial-risk-assessment-tool-nationwide/.

²² Id.

²³ Risk Assessment Fact Sheet, Correctional Offender Management Profiling for Alternative Sanctions (COMPAS) Pretrial Release Risk Scale – II (PRRS-II) (June 20, 2019), https://law.stanford.edu/pretrial-risk-assessment-tools-factsheet-project/.

²⁴ Risk Assessment Fact Sheet, Virginia Pretrial Risk Assessment Instrument (VPRAI) (June 19, 2019), https://law.stanford.edu/pretrial-risk-assessment-tools-factsheet-project/.

²⁵ EBDM Case Studies, supra note 7, at 39.

²⁶ Id. at 38.