Governance Design and Rehabilitation Efficiency in Small Private Health-Care Facilities: A Markov-Based Health Planning Model for Psychiatric Halfway Houses

Effective governance in psychiatric halfway houses requires a delicate balance between regulatory compliance (upward accountability) and patient-centered service quality (downward accountability). Although financial sustainability and formal compliance are often prioritized in practice, the role of downward accountability mechanisms in shaping long-term rehabilitation planning remains insufficiently examined. This study develops a strategic planning model to assess how institutional transparency and grievance mechanisms influence recovery trajectories among residents of psychiatric halfway houses. Using a Markov chain modeling approach, we analyzed state-transition probabilities across 21 psychiatric halfway houses in Taiwan between January and December 2023. Governance data were integrated from facility managers (n = 21) and staff surveys (n = 83) to operationalize three core dimensions of downward accountability: information disclosure, complaint procedures, and patient participation. The Markov analysis indicates that robust downward accountability is significantly associated with improved rehabilitation outcomes. In particular, higher levels of institutional transparency through information disclosure (p < 0.01) and structured grievance mechanisms (p < 0.05) were linked to increased recovery rates and successful community reintegration, whereas passive forms of patient participation showed no significant association with recovery outcomes. These findings suggest that sustainability in small-scale mental health facilities requires a shift from compliance-oriented governance toward patient-centered accountability structures. Health planners and accreditation bodies should therefore incorporate downward accountability indicators—especially transparency and complaint resolution—into standardized quality assessment frameworks. The proposed Markov-based monitoring model offers a practical, real-time decision-support tool to assist administrators in optimizing service delivery and enhancing patient-centered rehabilitation outcomes.