Study of US law enforcement co-responder programs identifies wide variations

In response to demands for police reform, agencies have begun to pursue alternative responses to calls involving mental health crises. Across the United States, jurisdictions are adopting co-responder teams that bring qualified mental or behavioral health professionals into police emergency responses calls. In a new study, researchers surveyed local and state law enforcement agencies to determine the prevalence and use of these teams. They found wide variation in the types of teams and how they operate.

The study, by researchers at George Mason University and the University of Wyoming, appears in Policing: An International Journal.

            “Our survey is the first to examine the prevalence and characteristics of law enforcement co-responder programs in the United States,” explains Cynthia Lum, professor of criminology, law, and society at George Mason University, who coauthored the study. “Our findings represent a first step in evaluating the effectiveness of these programs and in turn, developing evidence-based guidelines and protocols for their implementation.” Lum is an expert whose work is promoted by the NCJA Crime and Justice Research Alliance, which is funded by the National Criminal Justice Association.

Since the deinstitutionalization of state mental health hospitals in the 1960s, law enforcement has become the primary responders to people experiencing a mental health crisis. But the practice has come under intense scrutiny as a result of high-profile officer-involved deaths of people experiencing mental health crises. Among alternative responses is the co-responder model, which involves police partnering with qualified mental health professionals to respond to mental and behavioral health calls. However, the extent, nature, and effectiveness of these programs has not been widely investigated.

In this study, researchers developed and administered a survey to a nationally representative sample of local and state law enforcement agencies. A total of 568 agencies from all regions of the United States completed the survey between May 2022 and May 2023; 70% were local, county, or regional police departments; 24% were sheriff’s offices; and 3.5% were state or highway patrol agencies.

            The study found wide variations in staffing, operations, qualifications, and characteristics of co-responder programs, as well as in respondents’ beliefs about the effectiveness of such programs. The study also found a lack of consistent programmatic guidance to inform programs’ development, which results in personnel and funding challenges to implementing and maintaining co-responder programs. Specifically, the study found that:

• Nearly 88% of responding agencies had officers certified to be part of crisis intervention teams (CITs) and 41% reported having a co-responder program.
• Of the 232 agencies with co-responder teams, almost half (45.7%) started their program since 2020, 39% of the agencies had just one co-responder or qualified mental health professional to respond to calls with an officer (i.e., one team); 32% had two to three teams, 12% had four to six teams, and 9% had seven or more teams.
• Only 19% of responding agencies had co-responder teams available around the clock; 55% had teams operating on certain days and times, and 53% said when teams were not in operation, they were available to assist over the phone or by radio.
• In terms of training, 48% of agencies said co-responders received no additional training from the law enforcement agencies. Of the 48% that did provide training, the most common was maintaining personal and officer safety, followed by guidance on police procedures and culture, state laws, and use of force.
• Only 31% of responding agencies had some plan to evaluate or track the effectiveness of their co-responder teams by assessing outcomes such as use of force, arrests, officers’ time spent on calls, repeat calls, and hospitalization. Evaluation and tracking were sometimes tied to how programs were funded or managed (78% of programs were funded by local or state governments and 27% were funded by federal grants).

            “Our study suggests that several characteristics of co-responder programs are ripe for evaluation, including qualifications and training for mental health professionals and officers; protocols for implementation and operation; assessments of officer buy-in; understanding of data collection necessary for evaluating outcomes; and recommendations for building relationships with the community, between agencies, and with stakeholders,” notes Clair V. Uding, associate professor of criminal justice and sociology at the University of Wyoming, who led the study.

“While jurisdictions and communities are unique, efforts to better document programs’ development and encourage agencies to evaluate and share their information about effectiveness or lack thereof can help inform others who have adopted or want to adopt these programs.”

Among the study’s limitations, according to the authors, are that the high prevalence of CIT and co-responder programs among respondents could represent a selection bias; agencies without these programs may have disregarded the survey. In addition, the study was not designed to evaluate any specific aspect of co-responder programs but to provide information about agencies’ practices. And the study did not explore types of alternative responses to mental health crises that do not involve the police.

            “If we want more evidence-based approaches to co-responder units, police agencies’ approaches need to be tracked, assessed, and evaluated against desired outcomes,” adds Haley Moon, PhD student in psychology at the University of Wyoming, who coauthored the study. “Our results can help inform a theory of change framework for co-responder programs that can be applied to different communities with different needs and resources, to help identify short- and long-term goals, articulate required actions to achieve these goals, and develop measurable indicators of success.”