Predictors of hospitalization length of stay among re-admitted treatment-resistant Bipolar Disorder inpatients

Background Hospitalization accounts for significant health care resource utilization for treatment-resistant Bipolar Disorder (BD), especially among frequent users of acute inpatient psychiatric units. Appraisal of the clinical features and predictive role of selected variables is therefore crucial in such population, representing the aim of the present research. Methods A hundred and nineteen BD inpatients with an established history of pharmacological treatment resistance for either mania or bipolar depression were classified as long hospitalization cases (LOS+) and their controls and compared against each other for a number of demographic, clinical, and psychopathological features. Results Overall, female sex, current second-generation atypical antipsychotic (SGA)/mood stabilizer other than lithium as well as antidepressant treatment at the admission occurred statistically more frequently among LOS+ cases, concordant with higher scores at the Hamilton scales for depression and anxiety. Lithium utilization at the time of hospitalization did not differ between cases and controls (LOS-, n = 81/119), as predominant affective temperament and other psychopathological rating did not. Overall, the time of admission, use of SGA, anticonvulsant (other than lithium), antidepressant, lifetime alcohol dependence, and BD Type (-I or -II), but not current mood polarity at the time of hospitalization, correctly predicted LOS+ grouping 68.2% of the times: Exp(B) = 3.151, p042. Limitations Post-hoc, cross-sectional study, relatively small sample size, recall and selection bias on some diagnoses. Conclusions Overall, LOS+ treatment-resistant BD inpatients characterize for higher severity and greater pharmaco-utilization use, which warrants replication studies to include additional predictors to shed further light on the matter.