Town of Hamden

Police officers are currently asked to take on the roles of mental health specialists and social workers when responding to behavioral health crisis calls, working outside the policing skills emphasized in their training and ignoring their finite resources.¹ Unlike social workers who are trained to intervene during behavioral health emergencies, police officers often do not have access to the resources that address the underlying socio-economic and political determinants of health. By failing to acknowledge the root causes of these societal issues, traditional policing methods run the risk of upholding a system that perpetuates the criminalization of marginalized communities through law enforcement.²

Amidst recent social movements spurred by high-profile police brutality cases, municipalities across the nation have been urged to reimagine traditional policing practices. Community-police partnerships in the form of police-based specialized models, mental-health-based specialized models, and co-responder models have been tried over three decades in the United States in varying capacities and forms. The Town of Hamden, CT currently has a number of innovative alternative policing initiatives operating at the individual, neighborhood, and organizational level. Building upon this, the Hamden Police Department (HPD) in collaboration with the Keefe Community Center hopes to implement a streamlined, sustainable alternative policing model to foster community trust while addressing safety and health needs.

1. Conduct an environmental scan of alternative policing models to identify options and develop assessment criteria
2. Gain insights into community members’ existing relationships with the police and their perceptions of alternative policing models
3. Identify opportunities and challenges to implementation, including a cost-benefit analysis
4. Recommend next steps and evidence-based solutions to meet community safety needs

• Nine semi-structured interviews (45-60 minutes each) with key community stakeholders including behavioral health specialists, disability advocates, citizens and police officers
• Eight anonymous survey respondents from the Hamden Police Department
  • Open-ended interview and survey questions: role of the police, experiences with the police, social services, potential improvements, preferences, and feasibility of alternative policing models

• Financial data from Springfield Oregon, Eugene Oregon, and Hamden Connecticut’s budget archives were extracted from fiscal year 2009 to 2022 to examine potential cost savings through the implementation of a co-responder model.
  • A causal inference framework, specifically a differences-in-differences quasi-experimental design, was used to estimate the cost saving effect of a co-responder model by comparing Hamden and Springfield Oregon pre and post 2015 when the Crisis Assistance Helping Out On The Streets (CAHOOTS) co-responder model was implemented in Springfield Oregon

1. Interviewees felt HPD performed above average compared to other police departments, and identified room for improvement with regards to officer engagement and integration with the community. Participants ranked the HPD at an average of 7.4 on a scale of 1-10 with 10 being “the best a police department could be” and 1 being the “worst.”
2. Empathetic, non-threatening communication with individual officers has significantly strengthened relationships and sparked organic organizational partnerships. Conversely, ineffective communication in times of crisis can exacerbate existing distrust especially among those with mental illnesses or autism.
3. The politicization of policing was a common interview theme among both community representatives and police officers, and may have impacted participant responses. However, all stakeholders within and outside the department supported allocating resources to community services in an effort to assist officers and foster relationships, and felt this would be useful to both the department and community members.
4. A co-respondent model (police and mental health provider) was identified as preferable by 8/9 police respondents and all community stakeholders.

1. A co-responder model is projected to cost 2% of the annual police budget ($358,665 - $430,398) based on Springfield OR data and Hamden’s current budget.
2. Implementing a co-responder model resulted in an estimated 2.33% decrease in the total annual budget by diverting 17% of police calls, which translates to an average of $1,104,379 saved annually according to the differences-in-difference analysis.
3. Cost savings extend beyond the police department. Through the diversion of medical calls from Fire, EMS, and the ER, an estimated $3,908,333 is expected to be saved annually.

• Participants overwhelmingly supported a co-responder model where a police officer accompanied by a mental health or social worker respond to incidents together.
• Individual police-provider partnerships currently exist in Hamden and have demonstrated successful outcomes; However, resource limitations constrain expansion of these efforts.

• Create a comprehensive, streamlined programming plan that consolidates existing partnerships between behavioral health providers to optimize current resources.
• Use report results as a foundation to apply for federal grant funding as part of a 5-year financial transitional phase into a self-sustaining model.
• Offer training opportunities for officers to work with local community organizations to bolster the network of existing individual relationships.

• Strengthen coalitions between community leaders and police officers in tandem with outreach efforts to educate the wider community about initiative implications.
• Invest in community organizations to build mental health treatment and social service capacity using cost savings from co-responder model implementation.

• Interviews were limited to those willing to participate in this project despite community hesitancy around policing, Yale’s potentially contentious relationship with the community, and had access to video calling technology.
• The highly politicized nature of this research project complicated objective data collection though this was ameliorated through response deidentification and reassurance of confidentiality.
• Springfield and Hamden are distinct on multiple confounding factors such as demographic makeup and socioeconomic differences. The CAHOOTS model also benefited from existing community infrastructure. Therefore, cost savings figures should be extrapolated with caution.

1. International Association of Chiefs of Police. (2020, June 8). IACP Statement on “Defunding the Police.”
2. Kramer, R., & Remster, B. (2018). Stop, Frisk, and Assault? Racial Disparities in Police Use of Force During Investigatory Stops: Racial Disparities in Police Use of Force. Law & Society Review, 52(4), 960–993.
3. Hirschfield, P. J. (2015). Lethal Policing: Making Sense of American Exceptionalism. Sociological Forum, 30(4), 1109–1117.

1. Sewell, A. A. (2017). The Illness Associations of Police Violence: Differential Relationships by Ethnoracial Composition. Sociological Forum, 32, 975–997.
2. Loretta McNally. (2020). CAHOOTS Media Guide 2020.
3. Jackson Beck, Melissa Reuland, & Leah Pope. (2020). Case Study: CAHOOTS. Behavioral Health Crisis Alternatives.