A Closed Unit Incorporating Soteria Elements Versus a Conventional Closed Psychiatric Ward

After informed consent and completion of all baseline assessments, eligible participants are randomly assigned to either a small closed unit incorporating Soteria elements or a conventional closed psychiatric ward at the Jerusalem Mental Health Center. Treatment allocation remains concealed during the baseline assessment, so that participants and researchers completing the initial assessments do not know the participant's future treatment setting.

Because the study is conducted under real-world emergency psychiatric conditions, random allocation may not always be feasible because of clinical urgency, bed availability, administrative constraints, or difficulty obtaining consent and completing study procedures during an acute psychotic episode. When randomization cannot be implemented, eligible participants may still be enrolled and their method of allocation will be documented.

The closed unit incorporating Soteria elements provides care in a smaller setting of approximately 10 patients and emphasizes a calm and less institutional environment, supportive therapeutic relationships, shared decision-making whenever possible, personal autonomy, reduced reliance on coercive practices, and careful optimization of medication. The conventional closed ward provides standard acute inpatient psychiatric care in a larger ward setting.

The primary aims are to compare the relationship between antipsychotic medication dosage and improvement in psychotic symptoms, as well as therapeutic alliance and satisfaction with care, between the two treatment settings. Secondary outcomes include personal recovery, sense of dignity, internalized stigma, adverse events and coercive interventions during hospitalization, length of stay, and other clinical and post-discharge outcomes. Psychiatric readmission within one year after discharge is included as a primary outcome.

Baseline assessments are completed before treatment allocation is disclosed. At discharge and follow-up, full blinding is not always feasible because participants may reveal their treatment setting; however, outcome assessments will be conducted with partial blinding whenever possible, and the success of blinding will be documented.

The primary statistical analysis will focus on participants who were randomly assigned. Additional analyses will compare randomly and non-randomly allocated participants and will include all eligible participants using statistical adjustment for relevant baseline differences and potential confounding. Sensitivity analyses restricted to the randomized sample will be conducted to assess the robustness of the findings.