Abstract
Background Schizophrenia is characterized by frequent rehospitalization. We developed an individualized occupational therapy (IOT) program to facilitate proactive participation in treatment and improve outcomes. We evaluated whether the addition of IOT to group occupational therapy (GOT) as usual care, compared to GOT alone, during hospitalization would lead to a reduction in the rehospitalization risk of schizophrenia. Methodology We conducted a 2-year prospective cohort study following a randomized controlled trial comparing GOT+IOT and GOT-alone groups. Participants were schizophrenia patients discharged within 1 year from psychiatric hospitals across Japan. We utilized the Brief Assessment of Cognition in Schizophrenia (BACS), Quality of Life Scale (QLS), European Quality of Life 5 Dimensions 5 Level (EQ-5D-5L), Life Assessment Scale for the Mentally Ill (LASMI), Positive and Negative Syndrome Scale (PANSS), and modified Global Assessment of Functioning for Functioning (mGAF-F) at baseline, post-treatment, and 1 and 2 years after the index discharge. We comparatively analyzed the changes in each assessment from baseline to follow-up between groups using generalized linear mixed models. We used log-rank tests to compare the distribution of time to rehospitalization and plotted the Kaplan-Meier survival estimates for time to rehospitalization by group. We estimated hazard ratios (HRs) using Cox proportional hazards models to evaluate the impact of clinical factors on rehospitalization. We used Cox regression analyses to investigate the impact of each IOT component on rehospitalization. We conducted mediation analyses to evaluate the relationship between inpatient OT type and rehospitalization and the indirect effect of medication adherence on the relationship between inpatient OT type and rehospitalization. Results Among 66 who met the criteria, 34 were in GOT+IOT and 32 were in GOT-alone groups. Compared to GOT alone, GOT+IOT resulted in significant improvements in changes from baseline to 1 and 2 years in BACS, QLS, EQ-5D-5L, LASMI, PANSS, and mGAF-F. Overall rehospitalization rate was 43.55%; the GOT+IOT group showed a significantly lower rate with four rehospitalized compared to 23 in the GOT-alone group. Time to rehospitalization was significantly longer for GOT+IOT than for GOT alone. The Cox proportional hazards model showed that OT type (HR=0.03), resident support persons (HR=0.32), and medication adherence (HR=0.33) were significantly associated with rehospitalization. The univariate and multivariate Cox regression analyses on the impact of IOT components on rehospitalization revealed no significant components. The inpatient OT type had a significant total effect on rehospitalization (β=0.72). The direct effect of inpatient OT type on rehospitalization was significant (β=0.64), and the indirect effect of inpatient OT type on rehospitalization via medication adherence was not significant (β=0.08). Conclusions Adding IOT to standard care significantly prolonged time to rehospitalization, and receiving IOT during hospitalization reduced the risk of rehospitalization for schizophrenia. Significant factors related to each IOT component that affected rehospitalization were not detected. IOT reduces the risk of rehospitalization via improved medication adherence and has a strong effect as IOT itself reduces the risk of rehospitalization. This study provides evidence that IOT could prevent rehospitalization of schizophrenia patients.