Barriers and Strengths: Lessons Learned from a 1-Year Mixed Methods Program Evaluation

Background and purpose: This presentation highlights key lessons learned from the semi-structured interviews that took place as part of a larger mixed-methods evaluation of the Albany County Crisis Officials Responding and Diverting (ACCORD) program. ACCORD is an alternative-to-the law enforcement crisis response model that involves a team of Emergency Medical Service (EMS) professionals and masters-level mental health clinicians. While there was no single agreed-upon definition of de-escalation found in the literature reviewed for this project, the lessons learned from this evaluation provide some insight into understanding, defining, and operationalizing de-escalation.

Methods: The program evaluation utilized a mixed methods study design. This presentation will focus on the qualitative data derived from in-depth, semi-structured interviews. All interviews were conducted virtually using Zoom, with participants representing 1) ACCORD implementation team members, 2) ACCORD supervisors (Including an EMS supervisor and a mental health clinician) 3) and the ACCORD team members themselves. The interviews were conducted in two phases; Phase I approximately 3 months after the intervention began, and Phase II occurring approximately 12 months after the intervention began. Interviews were transcribed and coded independently by 2 members of the research team utilizing an adapted version of Applied Thematic Analysis. Following the 1st level of analyses, the researchers compared codes to check for congruence prior to finalizing the data code book. The research team also compared phase one and phase two interviews to highlight any growth or change in the program implementation and to inform future quantitative data gathering and analysis.

Findings: Key findings from the in-depth interviews highlight four themes, 1) the importance and strength of the interdisciplinary team model, 2) the ability of the model to provide needed mental health linkage and crisis service in an area that has been historically underserved, 3) the flexibility and adaptability of the model, and 4) the model's contribution to the operationalizing of ‘de-escalation’. There were 4 themes that aligned with areas of model improvement, 1) the need for ongoing engagement of the community, 2) the importance of ongoing interdisciplinary training and education, 3) the overall lack of mental health providers in these areas, as well as 4) the struggles of collaborating with multiple local and county-wide agencies (dispatchers, police, mobile crisis) etc. The lessons learned from this multi-agency program evaluation provide valuable insight to the ACCORD program, as well as to other jurisdictions contemplating implementing a multi-agency alternative to law enforcement responses to mental health crises.

Conclusions and Implications: Findings highlight the strength of the ACCORD program in providing flexible, adaptable, and highly needed resources to a historically underserved population. Findings also promote positive evidence to support alternatives to law enforcement responses to substance use and mental health crisis calls in the community. This presentation further emphasizes the power of utilizing a mixed-methods approach to program evaluation and research.