Developed by
Hallie Fader-Towe, Katie Herman, and Joseph Hayashi, the CSG Justice Center
Kevin O’Connell, Project Director, Data Driven Recovery Project (Kevin@oconnellresearch.com)

Background

The Council of State Governments (CSG) Justice Center and Kevin O’Connell¹ partnered with California’s Mental Health Services Oversight and Accountability Commission (MHSOAC) in the spring of 2021 to provide a series of four virtual workshops to help California counties apply the lessons learned from the MHSOAC-funded Data-Driven Recovery Project, in which Mr. O’Connell worked with 13 counties to develop concrete data strategies and plans for diversion. 

This workshop series was designed to help participants address various aspects of using and analyzing local criminal justice and behavioral health data, accounting for legal and technical considerations for data sharing and analysis designed to support policymaking. Participants were able to see how other counties have brought together different types of court data and learned how to do similar work themselves. There were four sessions, each of which included a general presentation on a court processing topic, followed by an opportunity for participants to collaborate with the trainers and other participants on using their own data. The sessions were: 

Session 1: Early Identification of People with Behavioral Health Needs 

Session 2: Mental Health Diversion Data 

Session 3: Incompetent to Stand Trial Data 

Session 4: Mental Health Courts Data

The first session featured a presentation by Jessica Yates, Business Analyst at the San Luis Obispo Sheriff’s Department, who spoke about how that county has been able to develop a data strategy to identify people with behavioral health needs in jail, how information is shared across departments, and how data can be used across different intercepts in the criminal justice system. 

The second session featured presentations by Tara Ames, Project Coordinator at the Siskiyou County Health and Human Services Agency, and Todd Schirmer, Division Director of Forensic/Criminal Justice at the Marin County Department of Health and Human Services. Ms. Ames and Mr. Schirmer discussed how they’ve created data strategies to successfully develop diversion programs in their counties. 

Session 3 featured a presentation by Tiffanie Synott, Supervising Public Defender at the Sacramento Public Defender’s Office, who discussed how and when to determine if a defendant should be evaluated for competency, how to understand the needs of this population, and how she has successfully developed diversion programs in Sacramento for this specific population. 

Session 4 featured a panel on Collaborative Courts, which included Neil Besse from San Diego’s Office of the Primary Public Defender; Matthew Dix from San Diego County’s District Attorney’s Collaborative Justice Division; Breawna Lane, Program Administrator at the San Diego County’s Behavioral Health Court’s treatment provider, Telecare Corporation; and Stephanie Tanaka from the Plumas County District Attorney’s Office. The panelists discussed hypothetical case vignettes to highlight different perspectives when considering cases for mental health or other collaborative courts. The session also featured a presentation by Kevin O’Connell about mental health court process and mental health court metrics. The session ended by summarizing the resources and key takeaways from the entire workshop series. Participants were encouraged to use the information from the workshop series to consider policy changes in their jurisdictions to improve diversion options for people with behavioral health needs. 

The training series was intended for practitioners who are key decision-makers for accessing and using local data (e.g., court, jail, county behavioral health, and homelessness data). The trainings were designed to help people in the field (judges, public defenders, prosecutors, clinical managers, social workers, etc.) develop practical ways to use the data they collect, but who may not have a framework for monitoring or connecting programs to larger county themes. Sixty-eight people registered for this training series from 21 counties across the state. The counties with the largest numbers of registrants were Sacramento, San Bernardino, San Diego, San Jose, and Santa Clara. 

Participants were asked to come prepared with one specific goal, such as developing a process for sharing mental health screening information with public defenders and developing processes for early identification of people with behavioral health needs, identifying the number of people eligible for mental health diversion in the jail, getting a unified view of people in the criminal justice system across the sequential intercepts and various court programs, or addressing processes related to the competency system.

How to Use This Manual

This manual was designed to help session participants develop, organize, and centralize their criminal justice data systems and strategies that focus on diversion for people with behavioral health needs. However, this manual can be used by anyone interested in improving data collection, analysis, information sharing, etc., regardless of participation in the workshop series. This resource can be used by court administrators; jail staff (management and booking); behavioral health staff, both in the jail and in the community; judges; defense attorneys; and prosecutors to help facilitate collaboration and communication between these groups. The manual can be used during in-person planning meetings or individually by staff. 

The manual is organized into five sections: Different Reasons for Data Sharing, Early Identification of People with Behavioral Health Needs, Court-Based Mental Health Diversion, Incompetent to Stand Trial, and Collaborative Courts. The manual also includes “Nuts and Bolts” resources and separate appendices that can be used to develop data processes and policies.

Different Reasons for Data Sharing

Behavioral health diversion, including mental health diversion under California Penal Code 1001.36, benefits from data sharing and linkage in two distinct ways: 

Operational data sharing allows data from different system actors to inform case-by-case decision-making. For example, this can occur when a mental health screen in the jail is shared with a public defender to identify a potential candidate for mental health diversion or when a comprehensive collaboration case plan is developed using information from needs assessments conducted by a behavioral health treatment provider and a probation officer.

Data sharing for policy analysis, in contrast, generally uses larger datasets to identify opportunities or trends, such as the total number of people in a jail’s average daily jail population who are likely to be eligible for diversion based on legal and clinical eligibility criteria. 

While the same data—in this case, data on people with mental illnesses in jail—may be used for both purposes, considerations related to protecting individual privacy and providing appropriate staff to analyze data will be different depending on how the data linkage is being used. While the workshop series mostly focused on data sharing for policy analysis, questions did arise about operational data sharing; these are addressed in this manual’s “Nuts and Bolts” section. 

Early Identification of People with Behavioral Health Needs

Jails are often the first place that data are collected related to mental illness or substance use.² Universal screening for mental health and substance use treatment needs at booking allows for the early identification of treatment needs and positions the jail to have data that will be useful for policy analysis.³ (See Screening and Assessment for Behavioral Health Needs in California Jails Workshop Manual for additional information.) 

An analysis of jail data can provide empirical information about the following questions: 

• What kinds of crimes are people entering the jail accused of? 
• How often do people enter the jail for non-new crimes, such as warrants, technical violations, and various holds? 
• How long are people staying in jail? 
• How are people being released from jail? Are they being connected to services? 
• What types of services are people being connected to? 
• Who is coming back to jail and for what reasons?⁴ 
• Is there a group of people who are “high utilizers” with distinct patterns of justice involvement?⁵ 
• Are there disparities in any of these measures when analyzed by race, gender, or age?

Court-Based Mental Health Diversion

Behavioral health diversion, including mental health diversion under P.C. 1001.36, creates “off-ramps” from various points in the criminal justice system to community-based treatment and supports.⁶ The Sequential Intercept Model provides a popular framework for identifying potential points for behavioral health diversion.⁷ 

Developing a shared understanding, or process flow, of how people will be identified for diversion eligibility is an important first step for data analysis related to diversion. This process flow should be illustrated as a diagram that shows the following elements: 

• Identification of diversion candidates 
• Review of whether candidates are appropriate for diversion 
• When candidates express their interest/agree to participate in diversion 
• Development of a treatment plan 
• A court order placing an individual on diversion 
• Process for successful completion of diversion, including any record clearance 
• Process for unsuccessful diversion, including a return to regular criminal case processing 

In particular, developing a process flow diagram can help policy-makers answer the following questions:

• Given characteristics of the average daily jail population, how many people are likely to meet different eligibility criteria? 
• What types of referral dispositions do you have for cases that have been referred for diversion? Are there increases in cases where the DA has declined, or where the client has declined? What percent of referrals are accepted for diversion? 
• What is the time to disposition for referrals, and is it within case processing time standards to complete assessments and court proceedings? 
• Are there disparities when analyzed by race, gender, or age between who gets referred to diversion and who gets accepted into diversion?

Incompetent to Stand Trial (IST)

The competency process in California impacts people who are generally unable to participate in their own defense and are unlikely to understand the implications of their cases pending before them. An increase in cases where doubt in a defendant’s competency is raised, as well as those found incompetent has created growing waitlists in county jails. Looking at the types of cases where a doubt is declared, as well as how far into a jail stay a doubt about a defendant is raised, can give a good indication of how a competency system is operating in a county. By gaining a better understanding of the risk factors for IST, as well trajectories, counties can intervene earlier and develop opportunities for diversion.

The passage of AB 1810 in 2018 (and amendments effective in 2019) authorized courts to divert people with eligible diagnoses and charges who were IST or at risk of becoming IST in an effort to reduce the number of people being sent to the California Department of State Hospitals (DSH) for competency restoration. Additionally, funding was made available through DSH to counties to divert people who are IST or at risk of being found IST on felony charges who had been diagnosed with schizophrenia, schizoaffective disorder, or bipolar disorder.⁸ The counties participating in the diversion pilot can assess the value of diversion for people at risk of becoming or found IST, as well as the impacts of diverting people in terms of jail bed relief and cost avoidance.

While competency to stand (CST) trial is often seen as parallel to court processes, communities should work together to add the CST process to their existing process flow. This will help identify opportunities for diversion stemming from the CST process and places where the CST process itself can be improved.⁹

In particular, analyzing a jurisdiction’s CST process flow can help policymakers:

• Develop a profile of the people who are being sent into the CST process to understand past justice involvement, homelessness, and collaborative court failures.¹⁰
• Understand the flow of these cases through assessment and 1370 determination, including where delays are happening and how to reduce events that delay disposition.
• Identify any inequitable outcomes by race, gender, or language preference.¹¹
• Identify opportunities and implications for misdemeanor IST restoration.

Collaborative Courts, Including Mental Health Courts

Since the late 1990s, when mental health courts became part of a national conversation, they have often been a first response to how the courts can improve outcomes for people with mental illnesses.¹² However, additional research on problem-solving courts, and drug courts in particular, has indicated the importance of reserving these programs, which usually entail significant treatment and supervision, for people with the greatest needs and the greatest risk of recidivism.¹³ 

The Judicial Council of California provides information about all collaborative courts in the state, including mental health courts. 

Collaborative court referrals, acceptances, and participation should also be part of a community’s process flow. Similarly, a discussion of eligibility criteria for collaborative courts should occur alongside eligibility for mental health diversion and IST to show how these different areas of the courts may interact with one another. Questions communities should ask in these discussions include the following:

• Who is eligible for and who agrees to participate in mental health diversion? How is this group different from the group participating in collaborative courts?
• How many referrals are made to collaborative courts by defense attorneys? What percentage of these referrals are being accepted?
• How long does the process take to move from referral to disposition?
• Are there differences in acceptance rates over time or within certain subpopulations?
• How does the use of sanctions and rewards impact changes in jail usage for accepted clients?
• How do long-term caseload changes impact the sustainability of a mental health court?
• How can the impacts of avoiding jail beds for people accepted into mental health court be estimated?

Nuts and Bolts

Each workshop session also included a “nuts and bolts” component to help participants develop and execute a strategy for analyzing cross-system data, building on the process used in the Data-Driven Recovery Project counties. 

Each session had its own process map to orient people to the distinct steps that happen before a defendant is accepted for diversion, as well as what happens after. For practitioners, creating reference maps like this can help define case processing time standards, look at processes from different perspectives, and agree on what constitutes diversion success. From an analysis perspective, the process maps can help in developing comparison groups as well as relevant metrics on who is moving through a diversion program.

The appendices of this workbook include resources to help communities conduct data analysis and process mapping as well as the following: 

A. Develop a data strategy that lays out how various data fields across systems can be merged and then used for many purposes. 

B. Understand governance for data linkage, including privacy and security considerations to ensure merging is an ongoing project, not a one-time effort. 

C. Identify common data elements in different agencies to then use in a flexible scalable way. 

D. Develop a merged dataset that can be applied to different programs and populations as defined by the agreements. 

E. Choose appropriate metrics and indicators of impact across justice processes.