Engaging Community Networks to Improve Depression Services: A Cluster-Randomized Trial of a Community Engagement and Planning Intervention

Abstract

This paper explores the effects of a group-randomized controlled trial, Community Partners in Care (CPIC), on the development of interagency networks for collaborative depression care improvement between a community engagement and planning (CEP) intervention and a resources for services (RS) intervention that provided the same content solely via technical assistance to individual programs. Both interventions consisted of a diverse set of service agencies, including health, mental health, substance abuse treatment, social services, and community-trusted organizations such as churches and parks and recreation centers. Participants in the community councils for the CEP intervention reflected a range of agency leaders, staff, and other stakeholders. Network analysis of partnerships among agencies in the CEP versus RS condition, and qualitative analysis of perspectives on interagency network changes from multiple sources, suggested that agencies in the CEP intervention exhibited greater growth in partnership capacity among themselves than did RS agencies. CEP participants also viewed the coalition development intervention both as promoting collaboration in depression services and as a meaningful community capacity building activity. These descriptive results help to identify plausible mechanisms of action for the CPIC interventions and can be used to guide development of future community engagement interventions and evaluations in under-resourced communities.

Keywords: Community engagement; Community of practice; Community-based participatory research; Depression care; Partnership networks; Quality improvement.

Conflict of interest statement

Conflict of Interest

No conflicts of interest to report.

Figures

Figure 1.. Sampling flow diagram for CPIC agency network analysis

Figure 2.. Agency network graphs at baseline and follow-up, CEP vs RS condition (any type of partnership interaction)a

a Includes any of five types of interactions (referrals sent, referrals received, joint client case management, program administration/funding, or public education/advocacy). Note: Each node represents an agency labeled with a unique study-assigned ID number. Node shape indicates the primary service sector of an agency as listed in the key.

Figure 3.. Emphasis on network-related goals from 3 CPIC project perspectives

Percentages are based on n=9 study documents, n=10 Steering Council participant interviews, and n=18 CEP participant interviews and focus groups that mentioned any network-related goal statements.