Introduction

As recently as 2023, nearly 50 million American teens and adults were living with a substance use disorder (SUD).^[1] Increased substance use has led to a high rate of drug overdose deaths. Between 1999 and 2022, 1.15 million Americans died from a drug overdose, one-quarter of which deaths occurred between 2018 and 2022.^[2] SUDs and drug use often co-occur with mental health disorders. Of the 58.7 million American adults living with any mental illness in 2023, 42.4% had used an illegal drug during the previous year.^[3] Of the 14.6 million people with severe mental illness in 2023, 51.9% had used an illegal drug in the previous year. Illegal drug use during the same period for adults with no mental illness was 20.6%.

Jurisdictions across the country are increasingly interested in adopting a public health perspective rather than a punitive criminal justice response toward substance use and mental health concerns. Diversion programs were established several decades ago to reduce justice system involvement, often at or after the point of arrest.^[4] While diversion programs reduced the number of people entering jail or prison, they still involved some components of the arrest and court system. More recently, many jurisdictions have implemented deflection programs, which take a more preventive approach to public health and safety.^[5] Deflection programs first emerged around 2011, building upon many lessons learned in the diversion field but applying the intervention prior to an arrest or court involvement. Since then, deflection programs have received both popular and legislative support nationwide. Given the recent emergence of deflection programs, it is important to examine how such programs are planned and developed.^[6]

The State of Illinois has invested in developing police-led deflection programs, including the Illinois Deflection Avenues Reconnection Team (DART). Program development began with action planning, a structured process in which a facilitator guides participants through the steps needed to design and initiate a program. Action planning seeks to foster a collaborative environment to plan social programs that achieve common goals and are sustainable.^[7] To help design DART, five in-person action-planning sessions were held. The sessions were attended by community members from the six counties that would be served by DART, including law enforcement, social service providers, and members of advocacy agencies. As part of this implementation evaluation, we conducted field observations, administered surveys after each action-planning session, and distributed a collaboration survey at the end of the process. Through our evaluation, we sought to answer the following questions:

• What transpired during the action-planning sessions?
• Who attended these sessions, and who was absent?
• How was the final action plan structured, and what did it contain?
• What was successful, and what could have been done differently to enhance action planning?
• To what extent were the planning group members collaborative?
• What barriers remained at the end of action planning that could impede the success of the program?

Since police deflection programs are relatively new and continue to evolve, examining how DART’s leadership team engaged in the initial planning phase is important for understanding the extent to which the process was collaborative, productive, and responsive to local needs, and for identifying strengths and gaps that could shape subsequent implementation and sustainability. Understanding how cross-sector partners plan and structure deflection initiatives may help policymakers and practitioners design future programs that are more collaborative, responsive to community needs, and sustainable.

Background

The DART program was developed through a partnership between the Illinois Department of Human Services (IDHS) and the Illinois State Police (ISP). IDHS funds the development and operations of deflection programs through the Cannabis Regulation Fund, established by the Cannabis Regulation and Tax Act (410 ILCS 705). IDHS contracts with Treatment Alternatives for Safe Communities (TASC) to support the deflection initiative. Within TASC, the Center for Health and Justice (CHJ) provides training and technical assistance during the planning and implementation phases of deflection programs. CHJ facilitates action planning, which helps programs define clear, measurable goals and identify strategies and steps to achieve them before program launch.^[8] Once a deflection program is launched, law enforcement officers in the program area make referrals to the program. These officers include members of local law enforcement, such as city police departments and county sheriff’s offices, and state police. ISP officers make referrals through their work in multijurisdictional drug task forces, which combine the expertise of local, county, and state law enforcement officers to address drug trafficking and its effects on communities. Police officers are the primary source of referrals to deflection programs and are responsible for providing warm handoffs of individuals to TASC deflection specialists. After a warm handoff, deflection specialists connect these individuals with appropriate treatment and services.^[9]

DART was the seventh deflection program to emerge from the partnership between IDHS and ISP and was designed to serve six counties in central Illinois: Douglas, Coles, Piatt, DeWitt, Vermilion, and McLean (Figure 1). Like other deflection programs, DART aimed to create pathways for police officers to refer community members in need of services or other assistance to deflection specialists. Deflection specialists would then follow up with these community members to determine which services might be beneficial and connect them to service providers in the community. At the time of this evaluation, the DART program – like several previously developed deflection programs – relied on local ISP drug enforcement task forces to lead the referral process. The three task forces partnering with the DART program were the East Central Illinois Drug Task Force (ECITF), Task Force 6 (TF6), and the Vermilion Metropolitan Enforcement Group (VMEG).

In the months leading up to the launch of DART, local law enforcement and behavioral health service providers developed the program through a series of action-planning sessions led by TASC CHJ. The purpose of the action-planning sessions was to bring together community stakeholders with the aim of forming a final action plan as part of the program’s early development. The action plan would articulate the goals and strategies of the DART program, provide a blueprint for deflection specialists’ activities, and outline the mode of collaboration between law enforcement and service providers. In total, five action-planning sessions were held: three in August 2023 and two in October 2023. All sessions were held in Champaign, Illinois. Ultimately, the action-planning sessions resulted in the development of an action plan that identified three desired program goals:

1. Increase available services in the area.
2. Utilize deflection referrals to increase engagement and local advocacy.
3. Use education and training to promote referrals from first responders.

Additional details of the action plan, including the strategies by which the DART team planned to achieve these outcomes, are discussed in the findings of this article.

The DART program was formally launched on July 31, 2024.

Figure 1 Map of IDHS-Funded Illinois Deflection Sites

Note. This map displays the locations of deflection sites as of August 2024.

Methodology

We used a mixed-methods approach to examine how stakeholders planned the DART deflection program during a series of action-planning sessions. The ICJIA Institutional Review Board (IRB) determined that this project qualified as an evaluation study and did not require IRB review for research involving human subjects.

Field Observations

To understand the action-planning process, we conducted 30 hours of field observations across the five action-planning sessions, held on August 8, 9, and 10, and October 11 and 12, 2023. One evaluation team member took detailed notes on session content, participant interactions, and the dynamics existing between facilitators and participants. During field observations, we followed ethnographic best practices while taking typed fieldnotes, which were reviewed for analysis.^[10]

Action-Planning Surveys

We administered a paper survey to participants at the end of each action-planning session to measure perceptions of engagement, collaboration, program sustainability, and potential benefits. The survey response rate varied by day, as shown in Table 1.

Table 1

Attendance and Survey Response Rates during Action Planning

Note. Attendance data were based on attendance sheets and observational counts.

The response rates on Days 4 and 5 were considerably lower than those on the first 3 days of action planning. This may have been due in part to the formatting of the action-planning sessions on those days, which condensed 3 days’ worth of content into 2 days due to scheduling constraints. Since action-planning surveys were administered at the end of each day, attendees may have been less likely to complete surveys on longer days. Several attendees also left those sessions early and, therefore, were not present when surveys were administered.

Questions varied from day to day because they reflected the specific topics covered in each action-planning session. The first day’s survey consisted of four items on the action-planning process, five on community partner engagement, and six on demographic characteristics. We chose to gather demographic information only on Day 1 to keep surveys short and maximize completion rates across multiple days. Given anticipated high attendance on the first day of action planning, we used the demographic information gathered on that day as a reasonable proxy for the general composition of participants. Day 2’s survey consisted of 11 items on the action-planning process and five about community partner engagement. Day 3’s survey included four items on the action-planning process and three on program sustainability. Day 4’s survey consisted of six items on the action-planning process and six on community partner engagement and collaboration. It included an open-ended question asking participants about whether they would have liked to invite additional partners to sessions. The final day’s survey included 10 items on the action-planning process and four on anticipated program sustainability. All questions used a four-point Likert scale. Complete responses to daily surveys can be found in Appendix A of the PDF version of this article.

We performed descriptive statistics on the daily survey items and the Inventory items using Microsoft Excel. For the daily surveys, we calculated means for each item. Surveys were anonymous, so we did not calculate changes over time for individual respondents. Instead, we compared average item responses across days. A majority (67-75%) of participants in each session were part of a 15-member “core group” of attendees who attended at least four of the five action-planning sessions. We interpreted the survey results alongside field notes to synthesize findings on the action-planning process and participant perspectives.

Wilder Collaboration Factors Inventory

Toward the end of the action-planning process, we administered a paper survey to measure collaboration among group members. The survey we administered was the Wilder Collaboration Factors Inventory (Inventory), a tool developed by Mattessich and colleagues that has been used extensively in program evaluation over the past two decades^[11] and validated, indicating that the survey items measure the intended constructs.^[12] We administered the Inventory on Day 4 of action planning and again on Day 5 to participants who had not completed it the previous day. The Inventory was administered to 18 participants in total, and we received 11 responses (eight on Day 4 and three on Day 5), for a 61% response rate. The survey took approximately 15 minutes to complete. The Inventory consisted of 44 items rated on a five-point Likert scale from 1 (Strongly disagree) to 5 (Strongly agree). The 44 items were grouped into 22 factors, each comprised of one to three items. We calculated mean scores for individual items and composite mean scores for each factor.

Evaluation Findings

Action-Planning Attendees

Facilitators

We observed the organization and facilitation of the DART action-planning sessions, which were facilitated by two members of the TASC CHJ team, including the team’s Executive Director and a Deflection Program Manager. Additional support for the action-planning sessions was provided by three subject matter experts (SMEs), two of whom were contracted by TASC CHJ to provide technical assistance based on their experience developing deflection programs elsewhere in the country, and the third of whom was a TASC Inc. Deflection Administrator who provided input on the work of deflection specialists.^[13]

Action-Planning Participants

Thirty-four individuals attended at least one action-planning session. Eight participants attended all five sessions, seven attended three or four sessions, and 19 attended one or two sessions. Between 16 and 20 individuals participated in each session, including at least 12 members of the core group who attended four or more sessions. Action-planning participants were affiliated with various community organizations, primarily law enforcement and behavioral health service providers (Table 2).

Table 2

Action-Planning Participants by Organization and Type

Note. This table covers action-planning participants who attended at least 1 day of DART action-planning sessions in August or October of 2023.

Figure 2 shows the number of participants in attendance at each session by organizational affiliation.

Figure 2 Local Action-Planning Participants by Day and Participant Type

Note. Data were drawn from field observations and attendance sheets. Participants may be counted more than once across days.

On Day 1, most action-planning participants were female, White, and non-Hispanic (Table 3). Most participants worked in the counties covered by the DART program, although several lived outside those counties. Participation from individuals with lived experience of substance use disorder was limited, and representation of some counties in the service region—particularly McLean County—was also minimal.

Table 3

Demographics of Local Action-Planning Participants

Note. The sample comprised 17 attendees who attended Day 1 of action planning.

Findings on Action-Planning Proceedings

We analyzed findings from our field observations, action-planning surveys, and collaboration inventory to identify key themes that emerged during the DART action-planning process.

Field Observations: Participation and Rapport

On the first day, 18 community members were in attendance. Of that number, 11 were behavioral health professionals, three were members of law enforcement, two represented a mental health advocacy organization, one was a public health education professional, and one was a leader in the faith community. By the end of the sessions, 12 of the 18 members had become part of the core group of high-attending DART action-planning participants.

Early discussions during the first session included disagreement among participants about language used in the introductory presentation on deflection. The presentation described how deflection programs act as a bridge between first responders and treatment services. During the presentation, several participants raised concerns that some language in the slides was stigmatizing because it implied that substance use is associated with criminal activity. One of the facilitators responded that the language reflected stigmatizing beliefs held by much of the public and was therefore acceptable. Several participants then argued that part of deflection’s purpose is to address stigma and that language commonly used by the public should not be repeated if it perpetuates stigmatizing beliefs. The facilitator subsequently presented a series of law enforcement statistics describing associations between substance use and criminal activity. During this portion of the presentation, some participants were observed fidgeting or avoiding eye contact with the facilitator.

Five people who had attended the first day’s session and had initially intended to attend all sessions were absent the following day. A service provider informed the group that members of his organization were disappointed with the language used in the presentation and with how their concerns had been addressed. According to that participant, those individuals chose not to return to the planning sessions and did not intend to participate further in the program development process.

Differences in perspectives among participants were also evident during discussions about program priorities. At times, participants used sharp language when discussing the goals of the deflection program. These exchanges reflected differing views about whether the program should primarily focus on connecting individuals with substance use disorders to treatment and recovery services or place greater emphasis on prevention, education, and addressing stigma related to mental health and substance use.

Two community members with lived experience of substance use disorder and treatment attended the second day of the planning process. Observers noted that these participants were not consistently integrated into program development discussions. When participants were asked whether anything was missing from the program thus far, one individual emphasized the importance of participants having a sense of purpose and belonging. During the session, a local representative shared details about these individuals’ experiences. Despite the facilitator’s efforts to redirect the conversation, some participants appeared uncomfortable during the exchange. The two individuals with lived experience did not attend subsequent planning sessions. In later discussions, several participants referenced concerns about how the perspectives of individuals with lived experience had been incorporated into the planning process.

Another dynamic that appeared to influence participant engagement was facilitation style. When a facilitator used a more directive approach to guiding discussion, some participants appeared hesitant to contribute. When a more coaching-oriented facilitation style was used—i.e., one that encouraged participants to elaborate on their perspectives and respond to one another—participants spoke more openly with both facilitators and other attendees. Over time, this approach appeared to support more open discussion among participants.

By the final day of action planning, participants were communicating respectfully and actively contributing to discussions about program development. Fourteen participants attended the final session, and 10 later participated in the program’s implementation and became members of the DART leadership team. Observations across the five sessions suggested that participant engagement increased over time as participants became more familiar with one another.

Field Observations: Integration of Impacted Community Members

One limitation of the action planning process was the limited participation of individuals with lived experience of SUD or prior involvement with deflection programs. As noted previously, two community members with lived experience attended one of the planning sessions but did not participate in subsequent sessions. Observers noted that their perspectives were not consistently incorporated into discussions on program development.

During the session they attended, one of the participants with lived experience was asked whether he would have contacted a deflection specialist when struggling with SUD. He responded that he would have been unlikely to make a self-referral to such a program. In later sessions, however, service providers and law enforcement participants discussed self-referral as a potential entry point into the deflection program.

These exchanges suggest that perspectives from individuals with lived experience were present only briefly in the planning process and were not consistently represented in later discussions about program design.

Field Observations: Establishing Community Needs

Participants appeared engaged when prompted to identify community problems that the DART program could address. They described several challenges, including limited resources and funding, a need for education to address stigma related to substance use and mental health, and what some participants described as “operational silos” among behavioral health service providers. When discussing barriers to addressing these challenges, participants debated the relative impact of different organizations’ work in the community. At several points, service providers discussed the need for additional funding to support services and staffing. Facilitators also noted that IDHS, which had funded deflection specialists, was reconsidering aspects of its funding structure and the extent to which it could provide financial support to service providers.

A subsequent discussion focused on strengthening coordination between first responders and service providers. Participants again highlighted limited staffing and gaps in available services. Providers expressed particular concern about connecting individuals referred through the deflection program to residential treatment for SUD, noting that few such facilities were located within the participating counties.

Participants also noted that no service providers or agencies from McLean County, the most populous county in the DART region, were present at the planning sessions. McLean County accounts for roughly half of the combined population of the six counties involved in the initiative and is home to several treatment centers and behavioral health services in the region. Participants discussed the importance of engaging organizations from McLean County in future planning and implementation activities.

The SME also suggested that local legislators could help secure sustainable funding for the deflection program, particularly for components not funded by IDHS, such as direct service provision.

Field Observations: The Final DART Action Plan

At the culmination of the 5 days of action-planning sessions, the DART team produced an action plan identifying three desired program outcomes and 11 strategies to achieve them:

• Outcome 1: Increase available services in the area.

  o Strategy 1: Increase the use of peer support and non-traditional services.

  o Strategy 2: Obtain a funder for new or enhanced services.

  o Strategy 3: Partner with established non-profits.

  o Strategy 4: Increase stakeholder engagement and local advocacy.

• Outcome 2: Utilize deflection referrals to increase engagement in services.

  o Strategy 1: Use peer support and case management.

  o Strategy 2: Advertise the program.

  o Strategy 3: Celebrate milestones.

  o Strategy 4: Increase access to virtual services.

• Outcome 3: Utilize education and training to promote referrals from first responders.

  o Strategy 1: Provide SUD and MH training via Mobile Training Units.

  o Strategy 2: Advocate for legislation to mandate SUD education for officers.

  o Strategy 3: Obtain police and EMS buy-in for the program.

Action-Planning Session Surveys

Table 4 presents average survey ratings from participants across the 5 days of action-planning sessions. Respondents rated items from 1 (Strongly disagree) to 4 (Strongly agree). While early responses reflected mixed perceptions of the process and limited partner engagement, attendees of later sessions rated highly their collaboration, participation, and community buy-in. Since attendance varied over action-planning sessions, ratings should be interpreted as group-level trends rather than changes among the same individuals over time.

Table 4

Average Survey Ratings across Action-Planning Sessions by Day

Note. Ratings are based on a 4-point scale from 1 (Strongly disagree) to 4 (Strongly agree). Participants varied across sessions; therefore, results reflect group averages for each day rather than longitudinal changes among the same respondents.

In the survey comments section, respondents identified several areas for improvement. On Day 1, respondents requested that more people be brought into the action-planning sessions. On Day 2, respondents asked TASC CHJ facilitators to provide more in-depth responses to their questions. On Day 4, many respondents suggested involving fire departments, EMS, and elected officials. Respondents also noted that first responder participation in the sessions seemed limited (Day 2) and that it was concerning that no one from McLean County was in attendance (Day 3).

Collaboration Inventory Findings

The Inventory was administered on Days 4 and 5 of the action-planning process. Appendix B of the PDF version of this article shows responses and means for each item, organized by factor. Based on the suggested use of the Inventory, we used a mean of 4.0 or greater to indicate an area of strength that does not need attention; a mean between 3.0 and 3.9 to indicate an area that warrants discussion; and a mean lower than 3.0 to indicate an area of concern that warrants attention.^[14]

For 15 survey items with the highest scores, the means ranged from 4.0 to 4.5, indicating areas of strength that do not warrant concern. There were an additional 24 survey items with mean scores ranging from 3.0 to 3.9, suggesting areas that may warrant discussion.

Three items received the highest means from respondents:

• My organization will benefit from being involved in this collaboration. (M = 4.7)
• I have a lot of respect for the other people involved in this collaboration. (M = 4.6)
• Everyone who is a member of our collaborative group wants this project to succeed. (M = 4.6)

While the Inventory responses do not fully align with the daily survey and observational findings from the early action-planning sessions, they do align with the satisfaction reported in later sessions, around which time the Inventory was administered. These results indicate that respondents reported generally positive perceptions of collaboration and mutual respect at the time the Inventory was administered, alongside concerns about funding and group membership. The high average on the item regarding respect for one another is consistent with observations from later sessions. These findings align with observations from later sessions in which a coaching facilitation style was used.

Two items received means below 3.0, indicating that these items warrant attention:

• Our collaborative group has adequate funds to do what it wants to accomplish. (M = 2.4)
• All the organizations we need to be members of this collaborative group have become members. (M = 2.6)

The low-scoring items on the Inventory also mirror low-scoring items from the daily surveys and points of discussion throughout the process. Respondents indicated a lack of sufficient funds, both in the community as a whole and for the program itself. Responses also reflected a perceived need to bring more community members and organizations into the collaborative group.

Discussion

This discussion explains the evaluation findings and connects them to existing research and the DART program’s current stage of development. We highlight key findings and, where appropriate, offer suggestions for programmatic refinement and guidance as the program continues to evolve.

Facilitator Style and Experience

The facilitators’ approach may have influenced participant engagement during action planning. Research on leadership and facilitation styles emphasizes the importance of building trust, recommends active, empathetic listening, and encourages the acceptance of vulnerability, authenticity, and transparency.^[15] One of the two facilitators demonstrated a more directive style by assuming an authoritative role, providing a clear structure, and steering the group firmly. The other facilitator adopted a coaching style, using questioning techniques and providing feedback and encouragement, which can be effective for team-building. For example, this facilitator expressed empathy with participants’ challenging work. During the sessions led by the second facilitator, participants appeared more open and engaged, which may have been due in part to greater comfort with a coaching style than with an authoritative facilitation style. Facilitators should be aware of different approaches and consider adapting their style based on group dynamics. In addition, we found that the SME from a rural community appeared more relatable to participants from similar rural counties in DART than the SME from an urban community. In these sessions, participants seemed more receptive to SMEs with experience aligned with the local community. Because participants may be more receptive to SMEs who reflect their communities and experiences, planners should choose facilitators whose backgrounds align with those of members of the groups they are facilitating.^[16]

Future Inclusion of People with Lived Experience

The two session participants with lived experience appeared to value self-referral less than service providers or law enforcement who attended the sessions, underscoring the need to include people with lived experience in planning. In contrast to assumptions expressed by service providers and law enforcement, research suggests that many individuals with SUD or mental health challenges face significant barriers to self-referral, including fear of stigma, distrust of law enforcement, and lack of awareness of available services.^[17] Without input from individuals navigating these challenges, action plans for deflection programs like DART may not fully reflect the need for alternative outreach models, such as peer navigators, community-based referral networks, or non-police intervention points.

Including people with lived experience in action-planning sessions might also help address logistical and systemic barriers to participation. For instance, rural residents sometimes struggle with transportation to treatment centers,^[18] while individuals with co-occurring mental health and SUD conditions may face eligibility restrictions or long wait times that discourage engagement.^[19] Yet rather than addressing such logistical and systemic barriers, DART’s action-planning session discussions tended to focus on service provider capacity and funding constraints. The inclusion of people with lived experience would likely result in a more robust understanding of the barriers to reducing treatment access and support the development of deflection programs that better meet community needs.

For these reasons, planners developing action-planning processes for deflection programs should consider incorporating the perspectives of individuals who may be directly impacted by their future program. In addition to including people with lived experience in traditional stakeholder meetings, program planners should consider multiple avenues for participation. These could include:

• Establishing an advisory group of individuals with lived experience (e.g., a group of people in recovery from an SUD) to provide structured input throughout program development, implementation, and operations.
• Conducting focus groups or listening sessions with potential program participants to identify barriers and refine engagement strategies.
• Partnering with a peer recovery organization that can serve as an intermediary between deflection specialists and individuals in need of services.^[20]

By actively incorporating the perspectives of those most affected, action-planning sessions for future deflection programs in the statewide initiative or similar programs in other locations may be better positioned to design more effective, equitable, and community-informed interventions.

Rural Social Services

Field observations and survey findings indicated a need for additional SUD treatment and recovery services in rural counties involved in the DART program. Rural communities can face numerous barriers, including a lack of interagency coordination and communication; limited resources and qualified personnel; insufficient capacity in hospitals to treat SUDs; transportation barriers; and stigma and confidentiality concerns.^[21] Several of these barriers, including a lack of coordination and insufficient treatment capacity, were highlighted during discussions among DART action-planning attendees.

Additional resources may be needed to further support deflection programs that serve rural areas, strengthen collaborative networks, and increase capacity for SUD treatment.^[22] The State of Illinois has already invested in supporting rural communities through such avenues as the Recovery-Oriented System of Care Councils. These councils create an infrastructure with resources to address SUD and currently cover two of the six counties involved in DART.^[23] However, our findings suggest that additional state and federal investments in SUD treatment and recovery support may be needed to address the distinct needs of rural communities.^[24]

Evaluation Limitations and Future Directions

The small sample sizes for the daily surveys and the collaboration Inventory limit the generalizability of our survey findings. Although 34 community members participated in at least 1 day of the action planning process, the Inventory was administered to only 18 attendees, and only 11 completed it. The low response rate may be due to the length of the Inventory (which took approximately 15 minutes to complete) and the timing of its administration at the end of a long day of program development, compared with the 5-minute daily surveys. Due to the anonymity of responses, it is unclear whether the 11 participants who completed the Inventory are representative of the 34 people who attended at least one action-planning session. Furthermore, collaboration findings may not accurately represent the attitudes and beliefs of the community as a whole. Some session attrition was observed, and dissatisfied community members who did not return for later action-planning sessions were not surveyed further. Survey responses from subsequent days of the process, including those to the Inventory administered on Days 4 and 5, may be affected by self-selection bias because they were administered predominantly to people who had already expressed a positive perception of the program.

We collected demographic data for action-planning participants only on Day 1. While this approach was efficient and minimized survey fatigue, it also imposed some limitations. Specifically, demographic information was not systematically collected on subsequent days, preventing analysis of whether participant composition changed over the course of the action-planning process. As such, our demographic findings should be interpreted as offering a snapshot of initial participation rather than a complete account of demographic representation across all session days.

The findings from the surveys and session observations were specific to the DART action-planning processes and the communities served by this program. Findings and recommendations from this evaluation may not apply to action-planning process for other deflection programs in different communities. Further, the limited engagement of people with lived experience during the action-planning process meant that critical perspectives were not included in this evaluation. In addition to limited participation from people with lived experience, no community members of Black or Hispanic race or ethnicity were involved in the DART action-planning process. The six counties in which DART operates have a combined population that is 86% White, but additional efforts could support recruiting service providers who serve and are representative of diverse communities in the region for future action-planning processes. Given the lack of attendance by McLean County representatives, there may be other barriers specific to that county that were not identified during action planning. Understanding the needs of McLean County and their implications for the DART program operations would require additional evaluation.

Conclusion

We evaluated the action-planning process of six counties in central Illinois that collaborated to form the DART deflection program. DART uses a police referral model to connect community members with substance use and mental health services. Over the course of 5 days of action-planning sessions, local service providers and law enforcement learned from one another and collaborated to develop a shared action plan for the DART program. The action plan was intended to guide the development of DART by addressing community needs, utilizing local resources, and promoting collaboration among partners. Our evaluation yielded a set of findings that may enhance action-planning processes for future deflection programs. First, we found that facilitation style appeared to influence participant engagement and buy-in. Participants appeared more receptive to a facilitation style that emphasized listening and fostering a supportive environment. Action-planning sessions for future deflection programs may benefit from the inclusion of facilitators trained in active and empathetic listening. Second, participants appeared more receptive to SMEs whose experience reflected participants’ rural communities. Program planners developing future programs might consider matching planning participants with SMEs who have experience working in similar communities. Finally, service providers in the DART program region identified insufficient resources for recovery services as a barrier that could affect program implementation. IDHS might consider helping deflection sites identify sustainable funding sources and strengthen the services that receive referrals from deflection specialists.