BACKGROUND: This study was designed to quantify adjunctive antidepressant use in bipolar depression and investigate whether the use of adjunctive antidepressants or presence of comorbid anxiety impacted the rate of psychiatric rehospitalization.

METHODS: Charts were reviewed for 377 adult patients with bipolar I depression discharged from Butler Hospital, a nonprofit psychiatric hospital in Providence, Rhode Island, from January 2008 to July 2011. Primary outcomes were impact of antidepressant exposure on readmission rates and time to readmission in the year post-discharge. Secondary analyses examined outcomes associated with anxiety, antidepressant class, and affective switch.

RESULTS: Of the 377 patients, 233 (61.8%) were prescribed an antidepressant at discharge and 144 (38.2%) were not. Only serotoninnorepinephrine reuptake inhibitor (SNRI) exposure, compared with nonexposure, was associated with greater readmission (74.4% vs 56.7%, P = .008) and shorter time to readmission (165 ± 145 vs 210 ± 153 days, P = .021). No antidepressant was associated with affective switch (P > .1). The presence of comorbid anxiety predicted readmission (66.4% vs 52.2%, P = .013), time to readmission (191 ± 153 vs 220 ± 153 days, P = .036), and affective switch (HR = 8.60, CI, 2.54 to 29.15, P < .001).

CONCLUSIONS: In this study, being prescribed an antidepressant for bipolar depression had no significant impact on psychiatric hospital readmission. SNRIs were associated with poorer outcomes, as was the presence of comorbid anxiety.