Spotlight on JMHCP: How Johnson County, Kansas, Used Data to Plan and Map a Systemwide Response

Each month, the Justice Center spotlights high-quality collaborative criminal justice/mental health initiatives that have received funding from the Bureau of Justice Assistance’s Justice and Mental Health Collaboration Program (JMHCP). Justice Center staff members ask the practitioners in these programs to discuss some successes and challenges they have encountered in the planning and implementation process. This month’s profile is from Johnson County, Kansas, a 2010 planning and implementation grantee.

Once officials in Johnson County, Kan., recognized that they faced a major problem with the growing number of people with mental illnesses in the justice system, they were presented with a significant dilemma: How do you design a systemwide response in the absence of reliable data that demonstrates what type of intervention will have the greatest impact?

Different stakeholders had divergent feelings about what was most appropriate. Some advocated for starting a mental health court. (Officials in Wichita had just started the state’s first mental health court through a JMHCP grant.) Others pushed for opening crisis stabilization drop-off sites for police to transfer custody of people with mental illnesses more quickly than current protocols allowed.

“We knew we were facing some serious problems with the growing number of people with mental illnesses in our system. What we didn’t know was what the numbers really were,” said Risë Haneberg, the county’s criminal justice coordinator. “We did not want to risk making bad decisions based on war stories or anecdotal evidence rather than hard data.”

Officials came together to develop a strategy for using data to inform the planning process. The county sheriff, the district attorney, the county’s chief judge, the chief public defender, the director of the county mental health agency, chiefs from the police departments of the county’s major municipalities, a volunteer from NAMI-Johnson County, and others formed the leadership team. (The group also engaged a state legislator, corrections officials, and database experts as advisors.) These leaders appointed staff members to serve on a work team, which managed day-to-day activities. They applied to the Health Care Foundation of Greater Kansas City to obtain a grant, and received staff support and guidance through United Community Services of Johnson County.

The group used a planning tool developed by the SAMHSA National GAINS Center, the Sequential Intercept Model, and took the name the Mental Health and Criminal Justice Intercept Project (the “Intercept Project”). Using the Sequential Intercept Model, team members mapped how people flowed through their respective systems and identified the number of people with mental illnesses within their system. These system maps provided the lens through which team members could look at the data and decide priorities.

Analyzing the Data

To conduct this data analysis, the Intercept Project surveyed the county’s 15 law enforcement agencies, worked with the probation department to count the number of people who were on some form of medication for a mental illness over the course of a single day, and dug into the county jail data.

The “black box” of jail data presented a significant challenge. The jail had numerous estimates of how prevalent mental illness was among jail detainees—it measured the number of people on psychotropic medication and the number housed in the jail’s special management unit—but neither provided a conclusive figure. Furthermore, the Intercept Project team realized it knew very little about this population. For example, the team didn’t have the data that showed the charges faced by this population, how long they stayed in jail, or their previous involvement with the criminal justice system. Without these data, planners believed they didn’t know enough about their target population to tailor a response.

To look at its jail data, county officials reached out to the Council of State Governments (CSG) Justice Center (which coordinates the Criminal Justice/Mental Health Consensus Project) for technical assistance. Researchers from the CSG Justice Center helped the Intercept Project team develop a “flag” in its criminal justice database so that intake social workers could identify anyone referred for mental health treatment. With this identifier, researchers could now compare people with mental illnesses to the general population of inmates. The Justice Center conducted a prospective study that tracked outcomes for individuals flagged as having a mental illness for periods of three and six months.

Researchers’ findings helped inform the Intercept Project’s planning. Once the mental health flag was in place, it was made visible to researchers, for whom the data could be “de-identified” from any specific individual to protect privacy. (The data were also available to those involved with developing treatment plans for individuals in the Johnson County Jail.) Investigators looked at the mental health cohort’s demographic characteristics, type of offense, prior criminal history, risk level, diagnosis, average length of stay, and access to services. This expanded set of data provided an excellent base of information to make recommendations for specific collaborations and interventions.

This process would ultimately yield a detailed plan that included dozens of recommendations for how to improve the response to people with mental illnesses across the entire justice system—when they encounter law enforcement, at jail intake, during their incarceration, at reentry, and while under community corrections supervision. Most of these recommendations considered ways to improve protocols and processes in the absence of additional funding. The plan was approved unanimously by the Johnson County Criminal Justice Advisory Council—which oversees the work of the Intercept Project—in November 2010 and released to the public on January 20, 2011.

The Intercept Project’s Recommendations

These findings enabled the project team to propose a series of changes and reforms, including specific collaborative initiatives. Some data helped them reject certain strategies under discussion when the planning process began. For example, initially many suggested opening a crisis stabilization drop-off site, but data from the law enforcement surveys showed that the relatively low number of mental health calls that law enforcement reported on the survey did not justify the expense. Similarly, data from the jail analysis showed that there were a number of options for improving outcomes for this population without needing to start a mental health court.

The group placed an emphasis on making recommendations that didn’t require additional funding. Most of the group’s recommendations could be implemented immediately, once stakeholders adopted the consensus report. These included (but were not limited to) the following:

• Based on the data, the Intercept Project decided to adopt new screening protocols at jail intake. They established a process to cross-reference criminal justice data with data collected by the mental health system to identify overlapping populations, which helps in classification and case staffing decisions, medication management, and psychiatric treatment. The data analysis also highlighted that the number of people referred for mental health services was much higher than the number deemed eligible for diversion from the jail. The county already had a jail diversion program, but it was limited to people with serious and persistent mental illness (SPMI), and went underutilized because few people were identified through screening and assessment as having SPMI. The county learned it was missing a critical opportunity to identify people eligible for diversion because of its screening practices.
• The county participated in a statewide pilot program to conduct risk assessments (using the Level of Services Inventory-Revised, or LSI-R) for a subset of inmates. However, the data from the risk assessments were not entered into the county’s criminal justice database. The Intercept Project concluded that ensuring risk assessment scores are entered in the database will improve the county’s ability to target supervision and treatment resources.
• They realized that individuals who were referred to mental health services within the jail made up a significant percentage of those who were frequently and repeatedly jailed. (For people with 20 or more bookings, those with the mental health referral made up 45 percent of misdemeanor bookings and 46 percent of felony bookings.) They would now assess these individuals for SPMI, and add repeated bookings as a criterion for eligibility for diversion. They would also conduct joint case planning between the jail and the individual’s mental health provider, community mental health center, or a family member/significant other.
• The Intercept Project also found that compared to national figures of women in jail, Johnson County women, particularly African-American women, were underrepresented as having a mental illness. (The national rate of women in jail with a mental illness is 31 percent; in Johnson County, it’s 23.3 percent, and 10.5 percent for African-American women.) This led to the conclusion that they needed to consider a new screening tool. They also decided to screen in a more private setting (once the county’s new jail, already in construction, opened), as they determined that women, particularly women of color, were likely underreporting the extent to which they suffered from a mental illness.
• They found that a large portion of the individuals with mental illnesses in their jail weren’t actually Johnson County residents. Of the 1,411 individuals flagged for a mental health referral between January and April 2010, only 817 (58 percent) were actually from Johnson County. Most were from neighboring Wyandotte County. Seeing the trends in the data, Intercept Project realized the jail needed to include Wyandotte officials in the planning process for an individual’s reentry.

The data also pointed them toward ways to maximize funding opportunities by supporting additional staff training and creating new positions. (Recognizing funding limitations, Intercept Project focused on recommendations that could be accomplished without additional funding.) To improve law enforcement responses, the Intercept Project recommended expanding CIT training to 20 percent of all officers within each individual police agency. With the data in hand, they applied for (and received) a JMHCP grant that financially supports a mental health professional who responds to calls along with police.

Sustaining the Effort

After the year-and-a-half planning process, Johnson County now has a host of options to improve responses to people with mental illnesses across the criminal justice system. The data have allowed the Intercept Project to prioritize these options based on the needs of the county and available funding. The county took advantage of local sources of funding (the grant from the Health Care Foundation of Greater Kansas City) and support (the organizing capacity of United Community Services of Johnson County), as well as the expertise and resources of a national technical assistance provider (the CSG Justice Center) and federal funding through the Bureau of Justice Assistance.

“When we started this, we didn’t necessarily realize how big an issue this was across the country,” said Risë Haneberg. “All we knew was that the best way to solve our problem and to get our systems speaking to each other was to first collect the best data we could.”

The involvement of high-level officials throughout the process—such as the local sheriff, the head of the county mental health system, and a state representative—helped balance what was possible with what an ideal systemic response to people with mental illnesses in the justice system would look like.

“The Intercept Project looked at the overrepresentation of people with mental illnesses in the justice system from the broadest possible lens,” said State Representative Pat Colloton (R-28th District). “By taking a systemwide view of things and basing our decisions on data rather than something you can’t put your hands around, we were able to identify a number of ways to improve people’s lives and make the system operate more efficiently. The likelihood of receiving funding to keep a program going is that much higher when there’s hard data demonstrating its need and its effectiveness.”