"Revolving door" and Bipolar Disorders: a retrospective study in an acute inpatient unit

The present retrospective study investigated clinical correlates of the revolving door (RD) phenomenon in a population of subjects affected by Bipolar Disorders (BDs).

Medical records of subjects with BDs admitted to a psychiatric inpatient unit over a 5-year period of time were retrospectively reviewed and clinical data were extracted into an electronic dataset. "Revolving Door Subjects" (RDS) were defined as those who presented three or more "Revolving Door Hospitalizations" (RDH) during twelve months. Features of RDH were compared with non-RDH in order to identify characteristics associated with RD phenomenon and possible risk factors for readmission. To explore predictors of RDH, a stepwise backword logistic regression model was built, including the variables that were significantly associated with RDH in the bivariate analyses.

In our sample of 176 subjects affected by BDs, 53 (19.9%) RDH were identified. In the RDH group, a higher prevalence of mixed episodes (p=0.029) and medical co-morbidities (p=0.004) was detected. Subjects with repeated hospitalizations were more often committed to psychiatric residential facilities at discharge (p=0.002). Treatment features related to RDH were represented by a higher prescription rate of atypical antipsychotics (p=0.030), benzodiazepines (p=0.001) and antidepressants (p=0.048).

Findings from the present study suggest that the early identification and treatment of medical comorbidities and specific clinical features of BDs may help reducing the RD phenomenon in this population of subjects.