*Result*: A locally driven juvenile co-responder program: Planning and implementation.
*Further Information*
*This study outlines the development and implementation of a law enforcement led, co-responder program that occurred in one Florida county from 2017 to 2021. The goal was to reduce involuntary commitment (i.e., Baker Act) examinations initiated by frontline officers. The program was embedded in the sheriff's office, designed for youth under the age of 18, and relied on licensed therapists to provide on-scene crisis de-escalation assistance to responding officers. First, we describe the strategic planning process which included a survey of sworn officers and dispatchers (n = 165), semistructured interviews with the chain of command (n = 30), and analysis of months of administrative records. Then, we report findings from 29 months of program implementation. Data describing 206 juvenile mobile crisis responses, youths' suicidal risk factors, and involuntary commitment decisions are presented, as well as monthly and yearly trends in recorded calls for service that involve a youth experiencing a mental health crisis and their dispositions (i.e., involuntary commitment or de-escalation). A reduction in the rate of involuntary commitment examinations made by frontline officers, timely on-scene arrival by a mental health professional, and clear alignment in suicide risk severity and response dispositions were some of the observed benefits of the program. We highlight the critical role that police mental health collaborations have in providing effective care for youth in crisis and emphasize the critical role that these initiatives will play in the future. (PsycInfo Database Record (c) 2026 APA, all rights reserved).*
A Locally Driven Juvenile Co-Responder Program: Planning and Implementation
<cn> <bold>By: Kristina K. Childs</bold>>
> <bold>Caitlin M. Brady</bold>
>
> <bold>Nicole Brenenstuhl</bold>
>
> <bold>Corrie Kindyl</bold>
>
<bold>Acknowledgement: </bold>Femina Varghese served as action editor.This work was supported by funding from the Bureau of Justice Assistance (planning phase, 2017-MO-BX-0047; implementation phase, 2018-MO-BX-0001).Kristina K. Childs played a lead role in conceptualization, formal analysis, funding acquisition, methodology, writing–original draft, and writing–review and editing and a supporting role in data curation and project administration. Caitlin M. Brady played a lead role in data curation, a supporting role in conceptualization, formal analysis, investigation, methodology, project administration, and writing–original draft, and an equal role in writing–review and editing. Nicole Brenenstuhl played a lead role in resources, a supporting role in conceptualization and writing–review and editing, and an equal role in funding acquisition and project administration. Corrie Kindyl played a supporting role in conceptualization, funding acquisition, project administration, and writing–review and editing.
The United States is currently experiencing a nationwide movement to alter the ways in which police intervene in crisis or mental health-related incidents (Balfour et al., 2022; Shapiro et al., 2015). The rationale for police intervention in crisis situations is based on two fundamental responsibilities of the police: protecting the safety and welfare of the community and protecting citizens with disabilities who cannot care for themselves (Lamb et al., 2002; Teplin & Pruett, 1992). These sometimes-conflicting roles require police officers to respond to incidents involving individuals in crisis (usually through 911 or other emergency communication systems) and quickly determine whether transport to a mental health facility (i.e., involuntary commitment), arrest, or remaining in the community best meets the individual’s immediate needs. The burden to make these life-altering decisions is often placed on the frontline officer with no assistance from trained mental health professionals.
Concerns about the preparedness of police officers to make these decisions during incidents that involve children and adolescents who are experiencing a mental health crisis have recently come to the forefront of policy and practice (e.g., Bureau of Justice Assistance, 2018; Federal Commission on School Safety, 2018). Police officers are often put into situations that require them to make assessments about a young person’s mental stability without guidance from trained mental health professionals or proper training about youth development, adolescent symptoms and disorders, or best practices for de-escalating crisis situations with young people (Fix et al., 2021; Janopaul-Naylor et al., 2019). There are several reasons why this is concerning. For example, it is estimated that mental health disorders are first diagnosed in late adolescence or early adulthood (Solmi et al., 2022), suggesting that many encounters between police and children under the age of 15 involve an undiagnosed mental illness. During these situations, youth and their caretakers may not recognize the need for mental health intervention, impeding the officer’s understanding of the situation and their ability to respond effectively.
In addition, strong neuroscientific evidence indicates that the brain is not fully developed until the mid-20s and that brain immaturity is inversely related to cognitive functioning and impulse control (Casey et al., 2008; Monahan et al., 2015). Adolescence is a period of change, characterized by risk-seeking and impulsivity, strong emotional reactions, and growing autonomy (Steinberg, 2010). Mental health difficulties during this time can exacerbate these characteristics, increasing the likelihood for problematic behavior such as delinquency, aggression, or substance use (Elliott et al., 2012). These externalizing symptoms can manifest into behaviors that elicit a response from the police (e.g., fighting, disorderly conduct). Decades of research shows that young people with mental health problems (e.g., anxiety, depression) have a higher probability of interaction with the police (Geller et al., 2014) and are overrepresented in the justice system (Underwood & Washington, 2016).
<h31 id="ser-23-1-73-d383e168">Police Mental Health Collaborations</h31>Due to the complexity of police encounters involving individuals with mental health needs, formal police and mental health collaborations have grown in recent years (Janopaul-Naylor et al., 2019; Kane et al., 2018). The Bureau of Justice Assistance (2018) broadly defines a police mental health collaboration (PMHC) as a law enforcement-based program that enables officers to respond appropriately and safely to people with mental illness. One type of PMHC is the coordination of joint response teams that include a police officer and a mental health professional. These programs, often referred to as co-responder teams or mobile crisis response teams, serve as an option available to police officers when responding to incidents that involve a mental health crisis and on-scene collaboration with a trained mental health professional. While joint response teams vary in their structure, the goals of these programs are to de-escalate crisis situations in a way that allows individuals to remain in the community (i.e., divert from the system) and receive services in a timely manner.
The benefits of joint, or co-responder, programs are demonstrated through prior evaluations of justice system dispositions (e.g., arrest, Kane et al., 2018), service engagement (Kane et al., 2018), cost-effectiveness (Semple et al., 2021; Waters, 2021), involuntary commitment transport decisions (Puntis et al., 2018), and emergency hospitalizations (Blais & Brisebois, 2021). Across diverse samples and implementation contexts, joint response programs have been shown to reduce out of home dispositions of calls for service (i.e., involuntary commitment transports, arrest) and increase access to mental health services (i.e., service referral, service engagement). Research has also found that satisfaction with police encounters is higher among adults when a co-response team is involved (Lee et al., 2015). The integration of such programs into community-wide mental health initiatives is considered a best practice by a number of federal organizations supporting interventions that address the mental health needs of individuals in the United States (e.g., American Psychiatric Association, Song et al., 2022; Bureau of Justice Assistance, 2018; Substance Abuse and Mental Health Services Administration, 2020).
Despite the growing popularity of co-responder programs, there is a lack of published studies on the implementation of these programs that are designed specifically for youth. That is, very little research addressing local jurisdictions’ implementation processes, screening and triage protocols, and programmatic outcomes for children and adolescents has been disseminated in recent years. This study addresses this limitation by describing the strategic planning process and descriptive findings from 29 months of implementation of a co-responder program, named the Juvenile Mobile Crisis (JMC) program, in one Florida county.
<h31 id="ser-23-1-73-d383e216">The Present Study</h31>Under the Florida Mental Health Act (1971/2009), a person may be taken to a psychiatric receiving facility for an involuntary examination if: (a) there is reason to believe that the individual has a mental health problem that substantially interferes with their ability to meet the ordinary demands of living, (b) the individual has refused or cannot consent to voluntary examination, and (c) there is a strong likelihood of serious bodily harm to self or others without care. Involuntary examination can last for up to 72 hr, then the individual is released, consents to inpatient services, or is involuntary committed via a formal petition. The Florida Mental Health Act (1971/2009) is more commonly known as the “Baker Act.” This act identifies police as one of three groups of professionals in the state of Florida with the authority to determine that a youth needs involuntary crisis intervention (circuit court judges and clinically certified care providers are the other two professions). The term “Baker Act examination” is used in the present study to represent the involuntary commitment examination that is addressed in this statute (i.e., transport to a mental health receiving facility for up to 72 hr of observation, assessment, and brief intervention).
To respond to the increasing rates of Baker Act examinations of youths in the county, a planning team was developed in February 2017 with representatives from the three organizations: the sheriff’s office, a local mental health agency, and the local university. The sheriff’s office is a full-service law enforcement agency that serves a medium-sized jurisdiction with a population of about 227,000 individuals across 350 square miles. At the time of implementation, the agency had over 1,200 full-time employees including 450 sworn officers. An important organizational characteristic that facilitated program implementation is that the sheriff’s office houses its own Department of Family and Youth Services which had been addressing youth mental health for several years through the implementation of a county-wide system of care. This system of care contributed to the recognition by community stakeholders that additional resources for officers responding to incidents that involve youth experiencing a mental health crisis were needed.
The mental health partner had been providing community-based assessment and counseling services to youth and families in the region for close to two decades. The organization specializes in working with families involved in the child welfare and juvenile justice systems and has been an established partner of the Department of Family and Youth Services leadership. The licensed therapists for the project were contractual, worked on-call, and housed in a centrally located office at the sheriff’s office. Last, a research team from the local university served as the evaluation partner. The research team was responsible for developing grant proposals, securing institutional review board approvals, analyzing programmatic data, and conducting continuous quality improvement assessments throughout implementation. The planning team collaborated on every aspect of program development and implementation. This included the development of all research protocols, interpretation of research findings, strategic planning, and outcome monitoring. There was also a JMC advisory board, which included representatives from the sheriff’s office chain of command, public school system, local psychiatric receiving facility, and youth and their parents. The advisory board provided continual oversight and final approval of all strategic planning and implementation activities.
The JMC program was designed to reduce the number of police encounters with youth that resulted in a Baker Act examination. Program protocols were based on best practices for reducing unnecessary involuntary commitments of youth (Bureau of Justice Assistance, 2018). The program was available on weekdays from 11 a.m. to 7 p.m. Program hours were determined based on a review of the 2016 and 2017 administrative data (collected during the planning phase). Using the day and time of the incident, we found that most incidents occurred Monday through Friday between the hours of 11 a.m. and 7 p.m. During this time, after determining that the incident involved a youth experiencing a mental health crisis, officers had the option to request the JMC program. Requests for JMC services were made via dispatchers. Once the on-call JMC therapist received the request from dispatch, they contacted the officer for information about the incident and then contacted the youth’s legal guardian to obtain consent. Upon arrival, they were debriefed by the officer and other involved individuals (e.g., parents, guidance counselors, or school administrators). The JMC therapist would then meet with the youth, talk to them about how they were feeling, and administer the Columbia Suicide Severity Rating Scale (C-SSRS). The C-SSRS is a well-validated screener used in various settings (e.g., emergency departments, schools, detention centers, outpatient mental health clinics) to quantify the severity of suicidal ideation and behavior (Posner et al., 2011, see description below).
After their assessment was complete, the therapists communicated with the responding officer about their assessment of the youth’s risk of suicidality and ability to safety plan. The JMC therapist, officer, and other available supports (e.g., parents or school counselors) collaborated to determine the course of action that was in the best interest of the youth. The first option was de-escalation and involved the youth remaining in the community and developing a safety plan that all JMC parties agreed with. The second option was the initiation of a Baker Act examination, which was based on concerns about the youth’s immediate safety. All JMC youth/families were referred to care coordination services by the JMC therapist, regardless of the outcome of the response. The purpose of this study is to describe the JMC strategic planning process and descriptive findings from the first 29 months of implementation. Four goals guided the present study:
>
The sheriff’s office was first awarded a planning grant in 2017. These funds were used to support the assessment of agency practices and development of the strategic plan (i.e., the planning phase, Goal 1). The sheriff’s office was then awarded an implementation grant in 2019. These funds were used for strategic plan implementation and evaluation (i.e., implementation phase). The methodology and results from the planning phase are presented first. Then, we describe the methodology and findings from 29 months of program implementation. The target population for the JMC program was children and adolescents experiencing a mental health crisis.
JMC Planning Phase
>
The present study took place in one county located in Central Florida. The youth population (under the age of 18) is estimated to be 98,300, and the median household income is $69,548 (higher than the state of Florida, $58,864; U.S. Census Bureau, 2023). From 2007/08 to 2016/17, the county experienced a 46% increase in involuntary examinations of youth (i.e., under the Baker Act criteria, Christy et al., 2018). The county also had a higher rate of involuntary commitment examinations of youth compared to the statewide average (173 per 100,000 in the county compared to 163 per 100,000 in the state) and a higher proportion of total involuntary commitment examinations were minors (24% in the county, 16% in the state).
Four objectives were accomplished during the planning phase. Objective 1 involved a comprehensive analysis of agency practices and professionals’ beliefs about responding to incidents that involve youth experiencing a mental health crisis. The results from Objective 1 were then used to develop a data-driven training plan for first responders that addressed youth mental health and available community resources (Objective 2, see Brady & Childs, 2023), a strategic plan for the implementation of a JMC program designed exclusively for the target population (Objective 3), and an outcome monitoring plan to track JMC outcomes and trends in mental health-related incidents postprogram implementation (Objective 4). Data collection for Objective 1 included a review of existing administrative data on all incidents that involved a minor who was experiencing a mental health crisis, a survey of sworn officers (i.e., deputies) and dispatchers (i.e., communication specialists), and interviews of the sheriff’s office chain of command (i.e., captains, supervisors, chiefs). A description of the participants, data collection procedures, and measures specific to each source of data are described below and the corresponding results are presented directly after each data source.
<h31 id="ser-23-1-73-d383e274">Administrative Data</h31><bold>Method</bold>
Participants
>
The administrative data included all incidents that involved a youth experiencing a mental health crisis and a law enforcement response from January 2016 to October 2017. Across the 22 month period, there were a total of 1,410 incidents. Just over half of the recorded incidents involved females, 68% involved White youth, and the average age was 15 (
Procedure
>
Descriptive analyses of the available administrative data from January 2016 to October 2017 (
Measures
>
Gender was recorded on the form as male or female. Race was recorded as White or nonwhite. Age was recorded numerically. Incident characteristics included the location of the incident (e.g., school, home), date and time of the response, and whether the incident involved one or more of the following: brandishing a weapon, threats of violence, weapons, or acts of violence or aggression (each individual items coded no/yes). Our primary variable derived from this form was a dichotomous variable representing whether the incident resulted in the initiation of a Baker Act examination (i.e., involuntary commitment) or de-escalation and referral to community-based services (i.e., remained in the community).
<bold>Results</bold>
The average number of incidents that involved a youth experiencing a mental health crisis and a law enforcement response was 64 per month (
<bold>Method</bold>
Participants
>
A survey was distributed to all dispatchers and sworn officers assigned to the patrol and school safety divisions (i.e., the two divisions that most frequently encounter youths). The original sample included 193 respondents. After removing individuals who did not provide valid responses to most survey items, the final sample included 165 respondents, including 49 dispatchers and 116 sworn officers (21 school resource officers [SROs] and 95 patrol officers). Close to 61% of survey respondents identified as male, 86% identified as White/Caucasian, and 11% identified as Hispanic/Latino. Just over one quarter (26%) earned a high school diploma or general educational development, 19% earned an associate degree, and 37% earned a bachelor’s degree or higher. The average age of respondents was 39 years old (range = 20–65 years,
Procedure
>
The survey was distributed via agency email in April 2018. The survey consisted of several questions that asked about interactions and experiences with youths experiencing a mental health crisis, knowledge about agency policies, and prior training specifically addressing youth mental health needs. The survey was open for 1 month with three reminders sent by division leaders. The response rate for the survey was 49%.
Measures
>
Several Likert-type questions asked respondents to rate how often they contact a mental health provider during encounters with youth in crisis (
<bold>Results</bold>
Descriptive analysis showed that 81% of dispatchers and 50% of sworn officers reported that they “never” or “rarely” contacted a mental health provider to assist with de-escalating the incident. One plausible reason for the low rate of contact might be related to the finding that over half of respondents agreed with the statement “the availability of community-based service-oriented options for youth is low in the county” (55%). Another scenario considered was the finding that only 44% of respondents reported feeling confident in their ability to “identify mental health symptoms among youth” suggesting that some officers may not be prepared to recognize the signs that a youth is in crisis.
Information about prior training indicated that 54% of respondents received prior training on “crisis de-escalation or stabilization techniques to use with youth” and 61% received prior training addressing available community-based services for youth. A much larger proportion of SROs reported prior training (84%), compared to patrol officers (65%) and dispatchers (41%). These findings are somewhat conflicting with the low levels of confidence reported. Only 44% of respondents felt confident in their ability to “know what to say to a youth who is experiencing a mental health problem” and 51% felt confident in their ability to “find appropriate resources/services that could help a youth who is having a mental health problem.” However, importantly, over a third of respondents that reported training either could not remember when the previous training occurred or reported that the training was 5 or more years ago.
When asked about formal policies/procedures for guiding decisions about youth in crisis, most respondents (70%) indicated that they were aware of at least one formal policy. Although, out of the 60 respondents who answered the open-ended question asking for a description of the policy, 72% referred to the Baker Act statute or the general orders (i.e., written agency policy) that address the Baker Act guidelines. Importantly, no formal policies/procedures related to contacting a mental health provider, assessing the severity of mental health or suicide symptoms, or referring youths to community-based services were mentioned by survey respondents.
<h31 id="ser-23-1-73-d383e366">Interviews of Agency Leadership</h31><bold>Method</bold>
Participants
>
A total of 30 semistructured interviews were conducted with the chain of command from the Communication (i.e., dispatch), School Safety (i.e., SROs), and Patrol Divisions in May and June of 2018. Participants reported an average age of 43 years (
Procedure
>
Semistructured interviews with the agency’s chain of command were conducted to assess stakeholder support and perceptions about program feasibility. All interview questions were developed in collaboration with the planning team. Interviews were conducted in-person with one of two team members present, audio recorded, and then transcribed by the research team. Interviewers were trained graduate students with prior experience collecting qualitative data. Team members and agency leadership did not know each other prior to the interview or have prior institutional knowledge about the interview topics (e.g., internal policies, stakeholder support). The average interview length was 20 min with interviews ranging from 7 min to 34 min. Participants gave consent to participate in the interview and, separately, to be audio recorded.
We used a general inductive approach to analyze the interview data, following the procedures described by Thomas (2006). This approach involves a systematic procedure for analyzing qualitative data that is guided by the goals of a program evaluation, but still allows the findings to emerge directly from the raw data. When applied to the present study, the goals of the interviews were to gain a better understanding of administrators’ perspectives on officers’ encounters with youth in crisis, needed resources, and support for the implementation of a co-responder program. As such, the coding categories were developed a priori, based on the research questions and the goals of the broader project (e.g., informal policies/procedures, preparedness, support for program, barriers to program). Each team member individually coded the 30 transcripts. The results of the coding process were then used to extract broader themes within the data, merging findings from multiple codes into one theme. The research team met to collaboratively review the thematic findings for consistency, name and define each theme, and ensure alignment with program goals. Due to the busy schedule of agency leaders, reliability checks were not feasible. However, when possible, triangulation with results from the first responder survey, administrative data, or available written policies was used. For example, interview responses about policies and procedures regarding involuntary commitment were triangulated with survey responses and written general orders. Responses addressing the availability of community-based options other than involuntary commitment were compared to survey responses of frontline officers to ensure alignment in these results. Importantly, these triangulation efforts did not change the thematic findings.
Measures
>
Interview questions addressed several topics that were broken into separate sections within the interview guide. First, we asked about challenges faced by the agency that involve children and adolescents. Questions included “what is the most challenging issue faced by your organization” and asked about the frequency that individuals under their supervision interact with youth who are displaying signs of mental illness or who are in crisis, their parents or guardians, and mental health providers (separate questions). Then, questions about agency policies were asked. These questions assessed awareness of formal, written policies or protocols that guide officers’ responses to a youth who are displaying signs of a mental health crisis and written interagency agreements that formalize the sharing of information between law enforcement and mental health agencies.
The last section of the interviews addressed stakeholder support for a youth-focused crisis response program supported by the agency. Example questions included “As a supervisor, would you support the use of agency resources to implement a police mental health collaboration, such as a youth mobile crisis unit” “do you believe that frontline officers within your unit would support a police—mental health collaboration, such as a youth mobile crisis unit” and “can you identify some barriers to implementing a police—mental health collaboration, such as a youth mobile crisis unit within your agency.” These questions were always followed up with “why” or “why not” so that respondents could expand on their beliefs about the feasibility of a crisis response program within the sheriff’s office.
<bold>Results</bold>
Several themes were noted from the interviews of sheriff’s office chain of command. The most apparent theme was liability, which was routinely mentioned by interviewees across topics including reasons for determining that a Baker Act examination was needed, barriers to implementing a PMHC, and barriers to consulting with a mental health professional. For example, one respondent noted “everybody is worried about the liability. They don’t really think of okay what’s in the best interests of the child, they think of well I’m liable if something happens.” Another respondent stated, “I hate to use the term very pro Baker Act, but I am very pro-protect yourself.” Concerns about liability were also addressed during responses about information sharing and communicating with mental health providers, oftentimes referring to concerns about violating Health Insurance Portability and Accountability Act laws.
The second theme was a lack of alternative options available to frontline officers during and after the crisis. One interviewee stated: “I think some people or some deputies look at it as an easier out sometimes, a Baker Act. I would have to say that is not always the best option but sometimes it is the only option.” Another administrator noted: “There’s not many strategies we have, for us it is kind of black and white. They threaten and hurt themselves, either with a parent or by themselves, they go to the hospital. There’s no discretion or any other options.” These statements reflect the common perception among interviewees that frontline officers were being forced to decide on the best disposition out of only two options: involuntary commitment examination or do nothing. As one lieutenant noted: “we would Baker Act them and 3 days later they would come back or cut themselves again. To have something in between the Baker Act and doing nothing would have been nice in those situations.”
The third theme was related to perceived organizational barriers to implementing a PMHC. While most commented that they would support a PMHC because of the additional option the program would offer frontline officers, there were concerns about the lack of resources and time and workload requirements. For example, the length of time it would take to contact a mental health professional, for them to arrive on-scene, and the length of time the officer would need to remain on-scene were all concerns. One respondent noted “… the only thing I could see is response time, how long that will tie-up the deputy on-scene, which if he has to wait 2 hr for a mobile response team versus a 15-minute drive to the hospital.” Another interviewee commented: “To be held up for that long, I see as an issue. I’m assuming this isn’t like a quick fix, you spend a little bit of time there.” Additional barriers repeatedly mentioned were staff shortages, funding availability, and buy-in from frontline officers.
<h31 id="ser-23-1-73-d383e401">Major Conclusions Drawn From Objective 1 of the Planning Phase (Study Goal 1)</h31>The major conclusion from this initial assessment (i.e., Objective 1 of the planning phase) was that there was a clear need to formalize additional crisis response options for frontline officers when they encounter a youth experiencing a mental health crisis. This conclusion was based on the number of youth-involved incidents that resulted in a Baker Act examination, the low number of officers that reported consulting mental health professionals when assessing a youth’s risk to self or others, and the inconsistent responses from the survey and interviews about formalized agency policies guiding officer decisions. Another consistent message was that frontline officers (survey), dispatchers (survey), and their chain of command (interviews) supported the agency’s efforts to expand the community-based options available to officers. These findings informed the JMC program strategic plan, which was developed by the planning team and approved by the JMC advisory board.
The plan sought to develop a localized training curriculum targeting youth mental health needs, de-escalation strategies, community-based resources available to officers, and the process for requesting JMC services (Objective 2 of the planning phase, see Brady & Childs, 2023). Related to this study, the plan also outlined the action steps necessary to develop and implement the JMC program. These steps included developing policies and procedures, formalizing information sharing and financial agreements, implementing evidence-based screening/assessment practices, and hiring/training staff (Objective 3 of the planning phase). The final objective of the strategic plan was the creation of an outcome monitoring plan to monitor program impact and inform quality improvement (Objective 4 of the planning phase). Objectives 3 and 4 served as the foundation for the next phase of the JMC program: implementation and outcome monitoring.
Implementation and Outcome Monitoring
> <h31 id="ser-23-1-73-d383e412">Characteristics of JMC Responses and C-SSRS Scores (Study Goals 2 and 3)</h31>
<bold>Method</bold>
Participants
>
Administrative data were used to monitor program success during the 29 months of implementation (August 2019 to December 2021). The unit of analysis focused exclusively on JMC responses. There were 206 JMC responses for 190 youth ages 7–18 (
Procedure
>
JMC therapists are required to complete a series of forms that were shared with the research team monthly. Therapists recorded the time the request was received, the time the therapists arrived on-scene, and the time the JMC response ended. A narrative of the incident was also completed by the therapist describing the location of the response, reason(s) for the JMC request, youths’ presenting symptoms at the time of arrival, parental involvement and communication during the response, and any justifications for the outcome of the incident (i.e., Baker Act examination or safety plan in the community). A “face sheet” included youth demographic characteristics, self-reported prior and current service history, and responding officer information such as name, assignment, and contact information. Finally, the C-SSRS scoring sheet was also part of the data shared with the research team. All JMC data were deidentified and assigned a study ID number that was unique to each JMC incident.
Measures
>
Data points used to monitor program implementation quality included JMC response times (minutes from the time the call was received by the on-call mental health professional to arrival on-scene) and response length (duration of the JMC response in minutes). We also tracked the location of the incident based on whether the response occurred at school, home, or another location in the community (e.g., park, restaurant). Characteristics of youth served by the JMC program included race/ethnicity (White, Black, Hispanic/Latino, or other), gender (male, female, or nonbinary), grade (first grade to 12th grade), age (in years), and current mental health service involvement (yes/no).
C-SSRS data were used to measure youths’ severity of symptoms and to assess its utility during Baker Act decisions. There are six questions on the C-SSRS, which measure suicidal ideation and behavior in the past 30 days. Two questions were asked to all JMC youths (i.e., have you wished you were dead, have you had thoughts of killing yourself). Youths that answer “yes” to having suicidal thoughts in the past 30 days were then asked three additional questions (have you thought about how you might kill yourself, have you had some intention of acting on them, have you started to work out the details of how you might kill yourself). The final question asks if the youth had “ever done or started to do anything to end their life.” All C-SSRS items are rated dichotomously (i.e., no or yes).
<bold>Results</bold>
Characteristics of JMC Incidents
>
Table 1 summarizes the characteristics of the 206 JMC responses. On average, it took the mental health professional less than 30 min to arrive on-scene (
>
><anchor name="tbl1"></anchor>
Most responses occurred at school (86%). During program implementation, the sheriff’s office served 34 schools and 32 (94%) had at least one response on campus. Most JMC incidents (62%) occurred at a middle school, 18% occurred at an elementary school. Only 9% of incidents that occurred at an elementary schools resulted in a Baker Act examination while 27% of incidents that occurred at a middle school and 29% of incidents that occurred at a high school resulted in a Baker Act examination, χ<sups>2</sups>(2,
Sixty percent of youths served by the JMC program were White and 58% identified as a female. Compared to student population data from the county’s school district (Florida Department of Education, 2023), the program served a greater proportion of White youth (47% of the student population were White) and females (49% of the student population was female). Thirty-one percent of the youths served by the program were involved in mental health treatment at the time of the incident (i.e., therapist, psychiatrist, or medication management). Bivariate analyses showed that JMC responses that involved females, χ<sups>2</sups>(1,
Correspondence Among Self-Reported Suicidal Risk Factors and Incident Disposition
>
Results from the C-SSRS indicate that JMC services and Baker Act decisions were being made for the right youth. Across all youths served by the JMC program, 83% said they had wished they were dead at least once in the past month, 76% had thoughts of killing themselves, and 43% had previously taken action to end their life. Chi-square tests of group differences were used to examine differences in C-SSRS scores across JMC response disposition (i.e., initiation of a Baker Act examination or de-escalation and safety planning). These results are displayed in Table 2. Statistically significant differences in C-SSRS item responses were observed, most notably among Questions 3 through 5. For example, 53% of responses where a youth reported suicidal intent resulted in the initiation of a Baker Act examination, compared to 5% of youths who did not report suicidal intent, χ<sups>2</sups>(1,
>
><anchor name="tbl2"></anchor>
<bold>Method</bold>
Participants
>
The unit of analysis focused on all sheriff’s office incidents that involved a law enforcement response to a youth experiencing a mental health crisis during the evaluation period from 2017 to 2021. These data include incidents that did and did not include a JMC response. All administrative data were extracted from the agency’s database by a planning team member and provided to the research team on a quarterly basis. There was a total number of 190 incidents in 2017, and in 2018, there were 103 incidents where a deputy responded to a youth experiencing a mental health crisis. In 2019, there were 146 incidents tracked, 145 in 2020, and 269 deputy responses to youth experiencing a mental health crisis in 2021.
Procedure
>
We continued to collect the incident data (i.e., form completed by the responding officer) allowing us to track monthly counts of incidents before and after implementation. We restricted our analysis to incidents that occurred during program hours (i.e., weekdays between 11 a.m. and 7 p.m.) for consistent comparisons.
Measures
>
Each incident was coded based on whether the response involved the JMC program. Incidents that did not involve the JMC program were coded non-JMC incidents. This group represents the comparison group as non-JMC response represent the agency’s traditional response to youths experiencing a mental health crisis. Incidents that involved a JMC response were coded as JMC incidents. This group represents the intervention group. This coding allowed us to track monthly and yearly counts across three distinct measures: total incidents (JMC incidents + non-JMC incidents), JMC incidents only, and non-JMC incidents only. Our primary outcome was whether the incident resulted in the initiation of Baker Act examination. As such, all three types of incidents (total incidents, non-JMC incidents, JMC incidents) were broken down into whether the incident resulted in the initiation of a Baker Act examination by a deputy (yes/no).
Unfortunately, the COVID-19 pandemic prevented the ability to make valid conclusions about the program’s impact on trends in county-wide incidents involving youth in crisis and Baker Act decisions. The program was implemented for just over 6 months when the county school system closed and the sheriff’s office changed some of their routine practices to align with federal mitigation guidelines (i.e., March 2020). In addition to these mitigation efforts, the negative and ongoing impacts of the COVID-19 pandemic on children and adolescents’ mental health have been well documented (e.g., de Miranda et al., 2020; Leeb et al., 2020). Many mental health agencies in the county switched to providing telehealth services for extended periods of time, adding another layer of potentially adverse effects of the pandemic on children and adolescents with mental health problems (Barney et al., 2020). As a result, it is impossible to disentangle fluctuations in monthly counts of incidents due to these exogenous circumstances.
However, two measures that remained useful across the study period were the total number of incidents per month (i.e., JMC incidents + non-JMC incidents) and the proportion of these incidents that resulted in a Baker Act examination. By relying on proportions, we were able to monitor whether the county was reducing the frequency at which police–youth encounters resulted in a Baker Act examination—the overarching goal of the JMC program.
<bold>Results</bold>
Figure 1 plots monthly counts of the total number of incidents that involved a youth experiencing a mental health crisis (i.e., JMC incidents + non-JMC incidents) and the total number of incidents that resulted in a Baker Act examination (i.e., JMC incidents and non-JMC incidents that resulted in a Baker Act decision). Several important observations from this time series should be highlighted.
>
><anchor name="fig1"></anchor>
First, the number of incidents that involved a youth experiencing a mental health crisis in the county started to decline in late 2017 through 2018 (prior to JMC implementation), although the proportion of incidents that resulted in a Baker Act examination remained consistent. This is represented by the overlap in the trend lines for total incidents (solid line) and total Baker Act examinations (dotted line) from 2017 to 2018. Second, before and after JMC implementation, the months with the lowest number of incidents are summer months (June and July) and holiday months (November and December). Although, the proportion of Baker Act examinations during these months is consistently high. Third, after implementation (gray vertical line), the total number of incidents that involved a youth experiencing a mental health crisis increased, then decreased during late 2020 (COVID-19 pandemic mitigation restrictions were in place during this time), and then increased in 2021. Despite these fluctuations, the proportion of incidents that resulted in a Baker Act decision was reduced after JMC implementation, regardless of the total number of incidents that occurred each month. This is represented by the divergence in the trend lines in most months after August 2019 (i.e., start of the JMC program). Chi-square tests of group differences also showed that 88% of total incidents prior to JMC implementation resulted in a Baker Act examination and 59% of total incidents after JMC implementation resulted in a Baker Act examination, χ<sups>2</sups>(1,
Table 3 summarizes annual incidents that involved a youth experiencing a mental health crisis from 2017 to 2021, broken down by whether the incident involved a JMC response or a traditional response (i.e., non-JMC incident). Three important points should be highlighted from these trends. The annual difference in the proportion of incidents that resulted in a Baker Act examination is large and consistently shows that the JMC program had fewer Baker Act examination decisions. For example, in 2021, 26% of incidents that involved a JMC response resulted in a Baker Act examination compared to 82% of incidents that did not involve the JMC team, χ<sups>2</sups>(1,
>
><anchor name="tbl3"></anchor>
Discussion
>
Overall, the JMC program produced the intended result, which was to reduce the proportion of police encounters with youths that ended with a Baker Act examination. Data from 29 months of implementation showed that requesting on-scene assistance from a mental health professional can be timely, improve access to community-based intervention options for youth and their families, and reduce involuntary commitment decisions. In addition, while the number of youths served by the JMC program annually increased, the number of youths served by traditional police response (i.e., non-JMC youth) showed smaller and inconsistent fluctuations. These trends suggest that the implementation of the JMC program expanded the intervention reach of the sheriff’s office, providing an additional option for police officers to use when necessary. This likely scenario is also supported by the general increase in total incidents after program implementation, but overall reduction in the proportion of total incidents that resulted in a Baker Act examination.
The use of the JMC program as an additional resource means that some officers could have been relying on JMC services in situations when the Baker Act criteria were not met but it was clear that the youth and family needed additional intervention beyond what the officer could effectively offer. Extant research indicates that adolescents experience negative emotions during interactions with police, potentially exacerbating the current crisis (Jackson et al., 2019; Jones et al., 2022). These feelings include fear, distress, anxiety, and anger. From this perspective, mobile crisis services provide a conduit for which responding officers can provide youth (and their parents) an opportunity to talk with a trained professional until the crisis is de-escalated, evidence-based assessment of suicidal risk factors, and referrals to community-based services. Essentially, extending the intervention reach of the sheriff’s office by providing effective screening/assessment, de-escalation strategies, and community-based referrals for youth in the community.
Another important finding is the alignment in C-SSRS responses and JMC dispositions (i.e., youth who reported more severe suicidal risk factors were more likely to experience a Baker Act examination). Such alignment suggests that officers did (in most circumstances) consider the input of validated screeners and mental health professionals when making Baker Act decisions. One advantage of incorporating a tool such as the C-SSRS into involuntary commitment decisions is the availability of these data to continuously examine aggregate C-SSRS scores and their alignment with JMC decisions as one proxy for program fidelity. Another implication of this finding is that validated screeners, like the C-SSRS, could serve as useful tools for officers to complete during encounters with youth (with or without a mental health professional). This is especially relevant to jurisdictions where PMHCs may not be feasible, but use of a quick and free screener is a feasible option for frontline officers.
The high reliance on JMC services at schools was not expected at the time of implementation. Retrospectively, however, there are several factors that likely contributed to wider utilization of the JMC program on school grounds. One obvious reason is that the role of the SRO is based on interaction with young people, thus increasing the likelihood of an encounter with a youth experiencing a mental health crisis. In Florida, the requirement to have at least one law enforcement officer in every public school (i.e., see SB 7026) means more opportunities for SROs to initiate involuntary commitment examinations of young people (i.e., due to the Florida Mental Health Act, 1971/2009). In districts where full-time mental health professionals are not available, the SRO can be the only available option for immediate crisis intervention. Prior research shows that SROs report feeling ill-equipped to respond to youth in crisis, inadequate training on youth mental health needs, and uncertainty about their role during crisis situations (Martinez-Prather et al., 2016). Based on this research, the higher-than-expected utilization in schools may reflect these sentiments among some of the SROs who requested JMC services.
Finally, the success of the planning and implementation phases serves as another example of the progress that can be made when police and mental health organizations work together with a common goal (e.g., Balfour et al., 2022; Shapiro et al., 2015). Importantly, buy-in for the implementation of a JMC program housed in the sheriff’s office was observed across frontline officers, dispatchers, and their chain of command. These findings align with several recent studies that demonstrate the universal desire among patrol officers for improvements to the resources available to them during interactions with individuals in crisis (Fix et al., 2021; Hassell, 2022). In addition, regular communication and feedback loops among planning team members, the use of data to communicate agency and JMC program needs effectively, and the securement of external funds from the Bureau of Justice Assistance to support the program were key drivers of successful implementation of the JMC program, which is still ongoing today.
<h31 id="ser-23-1-73-d383e628">Limitations and Future Research</h31>There are some limitations to this study that must be addressed. The first limitation is the lack of random assignment into the JMC or non-JMC groups. Unfortunately, random assignment to JMC was not feasible due to several reasons including concerns about fairness, issues with client confidentiality, and the rate at which the incidents occurred during program hours (i.e., small sample sizes). We accounted for this limitation by considering monthly trends in JMC and non-JMC incidents, although this is not a substitution for random assignment. The second limitation, discussed above, was the unprecedented impacts of the COVID-19 pandemic during both full years of program implementation. These factors prohibit any advanced statistical analyses such as interrupted time series analysis (McDowall et al., 2019). Without a clear understanding of the counterfactual, conclusions about program impact on county-wide incidents of police–youth encounters or youths served by the JMC program are not feasible.
The lack of C-SSRS data from the non-JMC incidents is also a limitation. Recall that officers had the option to request JMC services. This means that non-JMC incidents likely involved situations where the officer believed that JMC services were not necessary, either because there was no imminent mental health crisis or because the need for an involuntary commitment examination was clear and imminent. Since the JMC therapist is responsible for administering the C-SSRS, the use of this screener was not part of the traditional police response (i.e., non-JMC incidents). As such, it is impossible to compare the acuity of youths’ suicidal risk factors across incident types. Finally, our reliance on administrative data for outcome monitoring prohibited an assessment of youth’ experiences with the JMC program. Such data would have enhanced our understanding of the benefits and challenges experienced by program users.
Future studies on PMHCs should include evaluation designs that rely on random assignment, when feasible. These designs are needed to provide clear and convincing evidence of the efficacy of various PMHC program, including youth-focused co-responder programs. Longitudinal studies examining the longer term health and service utilization outcomes across youth who experienced a joint response and youth who experienced a traditional police response are also needed. Research should seek to understand how different types of police–youth encounters during crisis situations (e.g., mobile crisis, traditional police response) impact involuntary commitment stays, service engagement postcrisis, subsequent crisis events (e.g., calls for service), and client satisfaction. Although the JMC program was not originally designed to address equity concerns, future research should also examine whether co-responder programs are being requested for all youth in similar situations and that the dispositions of these responses are consistent across youth characteristics such as race, ethnicity, and gender.
Another imminent question for research and policy to consider is the role that crisis response programs should have in schools. While PMHCs are growing nationwide, so too is the placement of police officers in schools. The ambiguity about the SRO’s role in incidents involving students who are experiencing a mental health crisis becomes even more concerning given that 65% of schools in the United States reported having a security officer on campus, while only 42% reported mental health treatment services on campus (Irwin et al., 2022). Furthermore, a report conducted by the Southern Poverty Law Center (2021) noted that Florida has the highest rate of involuntary commitments of children in the United States and that many SRO-initiated Baker Act examinations are for behaviors that do not fall under statutory requirements (e.g., disobedience, mental health symptoms that can be addressed in the community). In addition, the report critiqued the role that police and school systems play in upholding Baker Act protections in Florida stating that school staff and police officers need “the resources and training to provide less-invasive, less-costly, and more-effective interventions that could help them better manage the behavior and needs of children in their care” (p. 2). PMHCs can serve as an effective strategy to balance the SROs role in mental health-related incidents and to increase SRO access to mental health professionals during crisis situations.
<h31 id="ser-23-1-73-d383e647">Conclusion</h31>The major takeaway from this study is the potential benefits of PMHCs for local police agencies and the youth they serve (i.e., reduced involuntary commitment examinations resulting from police–youth encounters). Our findings add to the growing body of evidence that co-responder programs lead to positive benefits for youth with mental health (Florida Mental Health Act, 1971/2009) needs and reduce organizational costs and resources (e.g., Semple et al., 2021; Vanderploeg et al., 2016). It is likely that the value of PMHCs will increase over the next decade, due to the rising rates of mental health disorders among children and adolescents (Bitsko et al., 2022) and the recent growth in the number of police officers placed in our nation’s schools (Irwin et al., 2022). As communities continue to grapple with shrinking financial resources, staffing shortages, and the rising rates of behavioral health needs among children and adolescents, the importance of developing strong PMHCs will become more apparent to police, mental health, and school administrators.
References
<anchor name="c1"></anchor>Balfour, M. E., Hahn Stephenson, A., Delany-Brumsey, A., Winsky, J., & Goldman, M. L. (2022). Cops, clinicians, or both? Collaborative approaches to responding to behavioral health emergencies.
Barney, A., Buckelew, S., Mesheriakova, V., & Raymond-Flesch, M. (2020). The COVID-19 pandemic and rapid implementation of adolescent and young adult telemedicine: Challenges and opportunities for innovation.
Bitsko, R. H., Claussen, A. H., Lichstein, J., Black, L. I., Jones, S. E., Danielson, M. L., Hoenig, J. M., Davis Jack, S. P., Brody, D. J., Gyawali, S., Maenner, M. J., Warner, M., Holland, K. M., Perou, R., Crosby, A. E., Blumberg, S. J., Avenevoli, S., Kaminski, J. W., Ghandour, R. M., … the Contributor. (2022). Mental health surveillance among children—United States, 2013–2019.
Blais, E., & Brisebois, D. (2021). Improving police responses to suicide-related emergencies: New evidence on the effectiveness of co-response police-mental health programs.
Brady, C. M., & Childs, K. K. (2023). Implementation and evaluation of a juvenile mental health training for law enforcement in a medium-sized jurisdiction.
Bureau of Justice Assistance. (2018).
Casey, B. J., Getz, S., & Galvan, A. (2008). The adolescent brain.
Christy, A., Rhode, S., Lersch, K., Ringoff, D., Jenkins, K., & Alitz, P. (2018).
de Miranda, D. M., da Silva Athanasio, B., Sena Oliveira, A. C., & Simoes-E-Silva, A. C. (2020). How is COVID-19 pandemic impacting mental health of children and adolescents?
Elliott, D. S., Huizinga, D., & Menard, S. (2012).
Federal Commission on School Safety. (2018).
Fix, R. L., Aaron, J., & Greenberg, S. (2021). Experience is not enough: Self-identified training needs of police working with adolescents.
Florida Department of Education. (2023).
Geller, A., Fagan, J., Tyler, T., & Link, B. G. (2014). Aggressive policing and the mental health of young urban men.
Hassell, K. D. (2022). Exploring officers’ perceptions of mandated crisis intervention training.
Irwin, V., Wang, K., Cui, J., & Thompson, A. (2022).
Jackson, D. B., Fahmy, C., Vaughn, M. G., & Testa, A. (2019). Police stops among at-risk youth: Repercussions for mental health.
Janopaul-Naylor, E., Morin, S. L., Mullin, B., Lee, E., & Barrett, J. G. (2019). Promising approaches to police–mental health partnerships to improve service utilization for at-risk youth.
Jones, N., Gius, B., Shields, M., Florence, A., Collings, S., Green, K., Watson, A., & Munson, M. (2022). Youths’ and young adults’ experiences of police involvement during initiation of involuntary psychiatric holds and transport.
Kane, E., Evans, E., & Shokraneh, F. (2018). Effectiveness of current policing-related mental health interventions: A systematic review.
Lamb, H. R., Weinberger, L. E., & DeCuir, W. J., Jr. (2002). The police and mental health.
Lee, S. J., Thomas, P., Doulis, C., Bowles, D., Henderson, K., Keppich-Arnold, S., Perez, E., & Stafrace, S. (2015). Outcomes achieved by and police and clinician perspectives on a joint police officer and mental health clinician mobile response unit.
Leeb, R. T., Bitsko, R. H., Radhakrishnan, L., Martinez, P., Njai, R., & Holland, K. M. (2020). Mental health–related emergency department visits among children aged <18 years during the COVID-19 pandemic—United States.
Martinez-Prather, K. E., McKenna, J. M., & Bowman, S. W. (2016). The impact of training on discipline outcomes in school-based policing.
McDowall, D., McCleary, R., & Bartos, B. J. (2019).
Monahan, K., Steinberg, L., & Piquero, A. R. (2015). Juvenile justice policy and practice: A developmental perspective.
Posner, K., Brown, G. K., Stanley, B., Brent, D. A., Yershova, K. V., Oquendo, M. A., Currier, G. W., Melvin, G. A., Greenhill, L., Shen, S., & Mann, J. J. (2011). The Columbia-Suicide Severity Rating Scale: Initial validity and internal consistency findings from three multisite studies with adolescents and adults.
Puntis, S., Perfect, D., Kirubarajan, A., Bolton, S., Davies, F., Hayes, A., Harriss, E., & Molodynski, A. (2018). A systematic review of co-responder models of police mental health ‘street’ triage.
Semple, T., Tomlin, M., Bennell, C., & Jenkins, B. (2021). An evaluation of a community-based mobile crisis intervention team in a small Canadian police service.
Shapiro, G. K., Cusi, A., Kirst, M., O’Campo, P., Nakhost, A., & Stergiopoulos, V. (2015). Co-responding police-mental health programs: A review.
Solmi, M., Radua, J., Olivola, M., Croce, E., Soardo, L., Salazar de Pablo, G., Il Shin, J., Kirkbride, J. B., Jones, P., Kim, J. H., Kim, J. Y., Carvalho, A. F., Seeman, M. V., Correll, C. U., & Fusar-Poli, P. (2022). Age at onset of mental disorders worldwide: Large-scale meta-analysis of 192 epidemiological studies.
Song, S. J., Balkozar, A., Frolov, L., & Gibbs, T. (2022).
Southern Poverty Law Center. (2021).
Steinberg, L. (2010). A dual systems model of adolescent risk-taking.
Substance Abuse and Mental Health Services Administration. (2020).
Teplin, L. A., & Pruett, N. S. (1992). Police as streetcorner psychiatrist: Managing the mentally ill.
Thomas, D. R. (2006). A general inductive approach for analyzing qualitative evaluation data.
U.S. Census Bureau. (2023).
Underwood, L. A., & Washington, A. (2016). Mental illness and juvenile offenders.
Vanderploeg, J. J., Lu, J. J., Marshall, T. M., & Stevens, K. (2016). Mobile crisis services for children and families: Advancing a community-based model in Connecticut.
Waters, R. (2021). Enlisting mental health workers, not cops, in mobile crisis response.