Communicating and Sharing Data with Judges During the Pretrial Process

This publication is designed to encourage more accurate and consistent data communication between pretrial agency staff and judges. In it, we provide examples of how staff can use program data to answer two key questions for judges:

1. Are we supervising people at the level we intended? Are there opportunities to right-size the caseload or adjust supervision conditions to better match our intentions?
2. Have recent pretrial decisions resulted in individuals on pretrial supervision successfully coming back to court and remaining arrest-free?

We encourage judges to share this publication with their pretrial supervision agency/division or court administration team and express to them a desire to review data more frequently to help guide decisions.

Communicating and Sharing Data with Judges During the Pretrial Process: A Guide for Pretrial Service Agency Staff

As part of the GPL’s Pretrial Initiative Technical Assistance, pretrial staff from jurisdictions across the country have shared that they face barriers in effectively and confidently communicating pretrial outcome data, such as the number of re-arrests or compliance with supervision conditions like GPS monitoring, to judges. Pretrial leaders need resources and solutions that help them develop stronger skills to analyze pretrial program data and communicate that data to judges in ways that are straightforward, intuitive, and actionable.

This publication includes specific metrics and types of data we recommend pretrial agency leaders start with to help answer two key data questions for judges:

1. Are we supervising people at the level we intended? Are there opportunities to right-size the caseload or adjust supervision conditions to better match our intentions?
2. Have recent pretrial decisions resulted in individuals on pretrial supervision successfully coming back to court and remaining arrest-free?

As we unpack each question, we provide step-by-step guidance on why it is important to be able to answer these questions and provide examples to help you get started working with your own data. We have also included examples of trends that pretrial leaders should watch for and tips for setting up a data-driven approach in their jurisdiction.

1. Are we supervising people at the level we intended? Are there opportunities to right-size the caseload or adjust supervision conditions to better match our intentions?

Why this matters: For every individual placed on pretrial supervision, pretrial staff must run several supervision checks. Staff often struggle to manage their caseload against the level of supervision needed for each client. In some jurisdictions the GPL has worked with, an individual pretrial officer’s caseload has included anywhere from 50 to 500 clients. In general, pretrial staff may conduct any or all of the following supervision services for each client, depending on the number and type of conditions they are assigned:

• Meet one to four times per month
• Run daily checks for new arrests
• Provide court reminders one to four times a month
• Monitor clients’ conditions between two to ten times a day
• Provide status reports to judges for each court date
• Submit violation reports
• Provide other case management support

If a judge assigns an individual to conditions that are mismatched to their risk levels, pretrial officers must still supervise them according to the conditions that the judge assigned. This can take away from pretrial officer’s ability to monitor more high-risk cases.

Suppose a judge has not seen the data on their overall condition assignment history. They may not be able to visualize patterns or trends such as over-assigning restrictive supervision conditions regardless of risk factor, which can result in an unmanageable caseload and an overall lower quality of pretrial supervision, see Examples A and B.

Getting Started with Data

1. Collect data on the total number of people on pretrial supervision, the total number of people with device conditions, and the average caseload ratio for pretrial staff on your team.
2. Review this data with the judges and consider whether that number is what you both Be prepared to also discuss whether your staff can adequately manage that many clients. See Example A for a list of high-level caseload stats that may be a good starting point if you are conducting this analysis for the first time.
3. If that number is not what you expected or is unmanageable, then calculate the proportion of clients by factors such as supervision level or condition type. You may also track how these data points have changed month to month. This will help you and the judge visualize the relative capacity of pretrial staff and understand if this is an ongoing change. See Example B for a helpful next step to take once you are more comfortable with your data or want to dig further into understanding your caseload.

Example A: Caseload Overview

Key Caseload Stats:

• Total number of individuals on caseload (as of today): 674
• Total number of individuals with device conditions (e.g., GPS monitor, alcohol monitors): 248
• Average caseload ratios (e.g., total caseload divided by number of officers): 1 officer: 89 clients.

Example Insight: Officers have shared concerns about their caseloads; on average, our officers supervise ~89 people at a time. They feel they cannot effectively supervise everyone on their caseloads.

Recommended Next Steps:

• Add context to high-level caseload stats by speaking with staff to understand how their caseloads and capacity are feeling.
• Share this overview with judges who set release conditions and discuss any capacity concerns.

Example B: Total Caseload by Condition Type Over Time

This graph answers the following questions: How many people are currently on supervision? What conditions do the individuals currently on supervision have? How has that changed in the last four months?

Key takeaways: The total caseload grew from ~160 individuals in March to over 300 individuals in June;~48% of the current caseload has a device condition.

Example insight: While the pretrial services agency’s total caseload has grown since March, the proportion of cases that require device monitoring, such as electronic and alcohol monitoring, has also increased. Between March and June, the percentage of cases requiring some type of device monitoring increased from 34% to 48%, while the overall caseload increased from 164 to 333. Both an increase in cases and an increase in device monitoring suggest that pretrial agencies’ workloads are increasing.

Data to illustrate how your caseload is growing could include:

• Growth in the number of individuals with conditions, despite a stable caseload indicating conditions are being used more frequently.
• Growth in the rate of condition placement alongside a growing caseload, indicating conditions may being used more frequently.
• A caseload size that is unmanageable for pretrial staff.

Diving Deeper into the Caseload Data

If you think your pretrial agency’s caseloads are becoming unmanageable, then you might consider using data to explore ways to right-size the caseload. Try to identify whether there are individuals who are under a higher level of supervision than necessary or who have a proven track record of returning to court and may no longer need supervision. These cases offer opportunities to right-size the caseload safely, rather than attempting to reduce it overall, which could have the opposite effect by potentially creating public safety challenges. Pretrial agencies can do this by collecting and reviewing data with your judges to answer the following questions:

How many days on average are people spending on pretrial supervision?

If people are on pretrial supervision for an extended period of time (e.g., more than 90 days) due to longer case processing times, pretrial agency leaders may consider speaking with a judge to highlight this data point and discuss reducing conditions for individuals with a track record of pretrial compliance.

How many people in my caseload have multiple conditions?

In some jurisdictions, individuals may be assigned multiple intensive supervision conditions at the same time, such as having to perform a remote breathalyzer and take a drug test. Using conditions in this way may be redundant, impose additional costs on the pretrial agency and time burden on the individual, and be mismatched to the initial charge. If pretrial officers observe that more than 20% of cases have multiple conditions or devices, agency leaders may consider reviewing these cases with a judge to remove one of the conditions.

What does my risk assessment data tell me about the risk level of individuals on different levels of supervision?

In some jurisdictions, pretrial agencies will make recommendations for release conditions or supervision levels based on a risk assessment. Ultimately, the judge decides the conditions under which the individual is released. In some cases, judges may assign individuals with a low risk score to a high supervision level as an additional safeguard. However, over-assigning conditions to clients over time may contribute to unnecessarily high caseloads for pretrial agencies.

If over 10% of cases that score low on the risk assessment are on high levels of supervision, this could be a signal to consider right-sizing the caseload by requesting that individuals with low risk scores either be placed on the lowest levels of supervision or taken off the caseload completely.

Example C: Quarterly Enrollments by Risk Score and Supervision Level

This graph answers the following questions: How many people were placed onto supervision this quarter? What supervision levels were they placed on? What were their risk scores? How do their risk scores compare to their supervision levels?

Key takeaways: Of the ~3,100 people placed on supervision this quarter, more than half were placed on high levels of supervision (Levels 3 and 4). Most individuals placed at these levels were also identified as low-risk. This indicates a mismatch between risk scores and supervision levels.

Example insight: Individuals with “low” risk scores are frequently being assigned to high supervision levels. Last quarter, 50% of all individuals on Level 3 Supervision and 66% of all individuals on Level 4 Supervision received a low risk score. There suggests potential overuse of pretrial supervision conditions, which may be contributing to a larger overall caseload without improving public safety or appearance rates.

Signs that supervision assignments are contributing to a bigger caseload: Low-risk clients being assigned to a high supervision level. These individuals could usually be released on their own recognizance or may require minimal monitoring. A growing caseload may be caused by a variety of factors. It may be a sign that more people were charged recently or that less people are in pretrial detention. These trends are not often within a pretrial agency’s control.

2. Have recent pretrial decisions resulted in individuals on pretrial supervision successfully coming back to court and remaining arrest-free?

Why this matters: Judges decide whether an individual can be released pretrial and, if they choose to release the individual, assign pretrial supervision conditions. Typically, judges make these release decisions using a combination of the individual’s risk assessment score (using a public safety risk assessment tool) and recommendations from pretrial services, along with other decision-making processes.

Judges have shared with the GPL that they are accustomed to receiving news of adverse client outcomes, such as noncompliance and rearrests. This perspective overlooks the broader context and can lead to the misconception that most individuals on pretrial supervision are not complying with release conditions. While judges may not have the capacity to review compliance reports for all individuals on supervision, providing overviews of court appearance¹ and safety rates² can help them to understand that pretrial supervision is often a safe option. This data can help judges build trust in pretrial supervision as a release option, increase their uptake of various pretrial release options, decrease their use of pretrial incarceration, and use less restrictive conditions more frequently.

Getting Started:

1. Collect data on the percentage of pretrial individuals who remained arrest-free and who attended all court hearings during their supervision period to share with the judge.
2. Segment this data over periods of three months and compare rates across at least six months of data.

Example D: Pretrial Outcomes Over Time

This graph answers the following questions: How many people made it to their court appearances and remained arrest-free? How has that changed from quarter to quarter?

Key takeaway: Pretrial outcomes have been relatively stable over the last six months.

Example insight: On average, pretrial outcomes, including safety rates and appearance rates, have been stable for the last two quarters. This could indicate that pretrial supervision continues to be a strong option even for individuals with a higher risk score, or who would have otherwise entered pretrial detention. This trend may also indicate that recent supervision assignments are having the intended impact on helping clients avoid re-arrest and supporting individuals to return to court for their case hearings.

Trends to watch for: Significant changes in the safety or appearance rate (five or more percentage points between two quarters).

Diving Deeper in the Outcomes Data

If your pretrial outcomes have been declining over the last six months, then you might consider using data to uncover additional patterns to help you understand why and be better prepared to explain those trends to a judge. Pretrial agencies can do this by collecting and reviewing data with your judges to answer the following questions:

• How do outcomes differ among different supervision levels or different reporting conditions?
• What have officers or case managers noticed about challenges their clients are facing?
• What supports have clients requested (transportation vouchers, court date reminders, etc.)?
• How do our safety and court appearance rates compare with other pretrial supervision programs’ rates? ³ 

Other factors to consider

Pretrial outcomes may naturally shift over time. For example, pretrial agencies we have worked with often report lower court appearance rates in the summer when parents have more childcare responsibilities and may find it harder to make it to court.

Pretrial outcomes may also decline due to factors within the pretrial system’s control, such as individuals being supervised at a level that is mismatched with their risk assessment score, program or policy shifts that have impacted outcomes⁴ , or because individuals have not been offered additional supportive services that would facilitate pretrial success, such as referrals for substance use counseling.

Four Strategies to Describe Trends and Discuss your Data with Judges

Using the GPL’s data-driven performance management (DDPM) framework, we have identified four communication strategies to help pretrial agencies translate program data on client progress into clear insights when communicating with judges:

1. Explain what the visualization or measurement is showing and emphasize each vizualization’s key takeaway.
2. Do not wait for the annual report; share data early and often.
3. Emphasize important implications from the data related to client outcomes or pretrial staff caseload management.
4. Identify action items based in the data for judges to improve client outcomes.

The following strategies are based on key takeaways collected during the GPL’s Pretrial Initiative’s Technical Assistance work in Harris County, Texas; Las Vegas, Nevada; Clark County, Washington; and Santa Cruz County, California.

1. Explain what the visualization or measurement is showing and emphasize each vizualization’s key takeaway.

Consider: What does a judge need to know about this information to trust it and use it to inform their behavior?

Judges may have questions about how pretrial staff came to certain conclusions, what the data are measuring, or where the data came from. Be prepared to answer questions about how you or someone on your team analyzed the data.

2. Do not wait for the annual report; share data early and often.

 Consider: How does the frequency of feedback provided to judges impact their ability to make decisions tied to client outcomes in real time?

Many judges have said that they appreciate frequent data sharing to provide regular feedback, reflect on progress, and understand the impact of their decisions on client success. Consider the difference between an annual report and a high-frequency, periodic report that a judge might receive from their pretrial division.

3. Emphasize important implications from the data related to client outcomes and outputs or pretrial staff caseload management.

Consider: Do the data reveal something your judge should worry about, or is everything operating as expected?

Judges may not be aware of how pretrial conditions impact client success. Therefore, it is important to clearly identify any implications from the data you are presenting. For example, if a greater proportion of cases in a specific court are assigned to a higher level of supervision, then that may result in an increase in the number of clients violating conditions overall. Similarly, assigning restrictive conditions to low-risk clients may increase the average amount of work pretrial staff must handle.

Other Frequently Asked Questions from Judges:

• Is this a concern or just a natural variation in data over time?
• Is this something within our control, and if not, who can impact this?
• Is this consistent with what we have seen in our jurisdiction before, or what we have heard from other places?

4. Identify action items for judges to improve client outcomes and support.

Consider: What are the specific actions a judge can take to facilitate better client outcomes?

You may identify several challenges with a client’s ability to manage and complete their supervision conditions. For example, a monthly drug test may require clients to pay a service fee, or a mandatory in- person check-in may result in clients missing work. However, many of these concerns are not immediately visible to a judge, as they are often more concerned with their responsibility to protect public safety. Here is a short list of common actions pretrial agencies may propose for judges to consider:

• Suggest that people with a low risk score are not placed on high or mismatched levels of supervision.
• Review conditions and supervision levels after 60-90 days on pretrial supervision to consider reducing supervision levels for some individuals.
• Suggest that judges consider adjusting the way technical violations are responded to and how they impact further condition placement.
• Explore whether assigning layered monitoring modalities is necessary or effective to maintain public safety (i.e., assigning drug testing and device conditions).

Footnotes

1. In the report, Measuring What Matters the U.S. Department of Justice National Institute of Corrections defines appearance rate as the percentage of released individuals who make all scheduled court appearances pending case disposition.
2. Safety rate is the percentage of defendants who are not charged with a new criminal offense pending case disposition. Many pretrial agencies establish success metrics using this report.
3. This report by the San Francisco Pretrial Diversion Project shows an example of how to compare pretrial supervision programs as a check to uncover potential patterns or differences among pretrial client outcomes.
4. In the National Conference of State Legislatures’ Pretrial Legislation Database, leaders can track recent policy changes made to the pretrial system, such as Virginia’s decision to require a human review any computer-generated risk assessments or Nevada’s decision to encourage enrollments in state-wide drug and substance use testing programs as a condition of expanding driving privileges for pretrial clients.
5. In the report, Measuring What Matters the U.S. Department of Justice National Institute of Corrections defines concurrence rate as the percentage of individuals whose court-ordered bail or release conditions correspond with their assessed risk level. Many pretrial agencies establish success metrics using this report.