Substance Use Severity, Autonomy, and Service Utilization Across Psychiatric Care Pathways in Romania: A Cross-Sectional Service Evaluation.

BACKGROUND/OBJECTIVES: Community psychiatric care aims to support recovery, autonomy, and lower reliance on inpatient services, but co-occurring substance use may reduce these gains. This study examined associations between psychiatric care pathway, non-tobacco substance-use severity, recovery-related patient-reported outcomes, clinical instability, and direct mental health cost in adults with severe mental illness in Romania. METHODS: This cross-sectional observational study included 128 adults receiving psychiatric hospital care, long-term residential care, or protected community housing within one regional psychiatric care network. Participants had at least 12 months of uninterrupted care in their current setting. Substance use was assessed with the World Health Organization Alcohol, Smoking and Substance Involvement Screening Test (ASSIST), excluding tobacco from the analytic severity score. Outcomes included SF-36 physical and mental scores, brief service-experience ratings of autonomy and perceived coercion, emergency department visits, relapse, readmission, inpatient days, outpatient visits, and annual direct mental health cost. RESULTS: Any non-tobacco substance use was most common in hospital care, intermediate in residential care, and least common in community housing. Mental quality of life was lowest among hospital participants with substance use and highest among community participants without substance use. In multivariable models, ASSIST-derived severity was independently associated with a lower SF-36 mental score and a higher direct cost, while autonomy was independently associated with better mental quality of life and lower direct cost. The cost outcome was modeled primarily with a gamma generalized linear model with a log link because costs were positive and right-skewed. Because pathway allocation was clinically determined rather than randomized, and because autonomy, perceived coercion, and mental quality of life were all self-reported at a single assessment, every multivariable and path estimate is exploratory. CONCLUSIONS: Community psychiatric pathways were associated with more favorable recovery indicators, but non-tobacco substance use remained an important marker of poorer outcomes and greater service burden. The findings support addiction-sensitive community psychiatric care and routine monitoring of autonomy and perceived coercion as service-quality indicators. Because the design was cross-sectional and the care pathways were non-equivalent at baseline, these results are hypothesis-generating and do not establish that care pathway, substance use, or autonomy exerts a causal effect on any outcome measured.