
What Can We Learn from How We’re Connected? Mapping Collaboration in Mental Health and Housing Systems
Abstract
Background: Integrated care depends on strong, effective relationships across the organizations that support people’s health and well-being. In areas like mental health, addiction, and housing, collaboration is both essential and complex, shaped by how partners share information, coordinate care, work toward common goals, and learn from one another. Understanding how these relationships function is a critical foundation for improving outcomes and building systems that learn and adapt over time.
Evaluating these connections also helps reveal where coordination breaks down, where communication can be strengthened, and how shared accountability can be fostered across sectors. The purpose of this study is to explore interorganizational collaboration within a regional integrated mental health and housing network in Ontario, Canada, to inform ongoing evaluation and guide future system improvement.
Approach: This exploratory study uses social network analysis to map collaboration across a regional integrated care network in Ontario focused on mental health and housing. Guided by D’Amour et al.’s Structuration Model of Collaboration, the Network Collaboration Survey was co-developed and refined through consultation with organizational partners to ensure alignment with local integration priorities. The survey examined constructs such as shared goals, trust, governance, leadership, and information exchange, and was validated through multiple rounds of expert review.
Organizations providing mental health, addiction, and housing services were invited to participate, resulting in a whole-network design capturing both formal and informal partnerships. Using UCINET 6 and NetDraw, analyses will examine structural and relational characteristics of the network, including tie strength, reciprocity, and homophily, while node-level metrics such as degree and betweenness centrality will identify key actors, bridging roles, and collaboration gaps across sectors.
Results: Analysis of this research is currently underway. However, preliminary mapping and analysis indicate several emerging patterns within the network. Collaboration across mental health and housing organizations appears strong, yet connections to primary care remain limited. Organizations that interact more frequently report higher levels of trust and information exchange, suggesting that relationship intensity supports stronger collaboration. Conversely, hospitals appear less connected within the broader system, particularly as perceived by community-based housing partners, who report lower trust and reduced information flow.
Preliminary visualization suggests a moderately centralized structure, with a few organizations serving as key bridges between otherwise disconnected clusters. These patterns highlight opportunities to strengthen formalized tools, shared protocols, and collective learning processes to promote more consistent collaboration. Ongoing analyses are exploring how organizational characteristics, such as size, sector, and information-custodian status, shape these dynamics. Early findings are being shared with system partners to inform ongoing capacity-building and integration efforts within the local health and housing system.
Implications: By revealing how organizations across the mental health and housing sectors are connected, this study provides a foundation for strengthening collaborative capacity within a regional integrated care network. The emerging baseline offers system leaders and partners actionable insights to target relationship-building efforts, enhance trust, and develop more inclusive mechanisms for information sharing and joint decision-making. These findings contribute to broader efforts to build an integrated, learning-oriented workforce capable of addressing complex needs through coordinated, cross-sector care.
© 2026 Amber Gillespie, Andrew Papadopoulos, Zvonimir Poljak, Carly Whitmore, published by Ubiquity Press
This work is licensed under the Creative Commons Attribution 4.0 License.