OBJECTIVES: To compare the risk of hospitalization and associated inpatient psychiatric treatment costs among patients with bipolar disorder treated with aripiprazole, ziprasidone, olanzapine, quetiapine or risperidone.   METHODS: A retrospective cohort study using the Pharmetrics Patient-Centric database from January 1, 2003 -September 30, 2008 was conducted.  Included patients were aged 18-64 years, diagnosed with bipolar disorder, and newly treated with atypical antipsychotic medications.  Patients were followed from treatment initiation to psychiatric hospitalization, medical hospitalization, discontinuation of index medication, loss to follow-up, or one year.  Mean length of therapy was similar among the treatment groups.  Generalized gamma regression compared inpatient costs between the groups.  RESULTS: Among 19,176 bipolar patients, those treated with aripiprazole (n=3,690) had a lower risk of hospitalization during their time on treatment (4.9%) compared to patients receiving ziprasidone (9.2% p<0.001, n=1515)), olanzapine (7.0% p<0.001, n=3038)), quetiapine (7.3% p<0.001, n=7936), and risperidone (7.2% p<0.001, n=2997).  Fewer hospitalizations translated into lower per-patient-per-month psychiatric-related inpatient healthcare costs for patients receiving aripiprazole ($121) compared to those receiving other atypical antipsychotics (ziprasidone $249 p<0.001, olanzapine ($203 p<0.001), and quetiapine ($182 p<0.006).  Differences in hospitalization costs were not statistically significant between aripiprazole and risperidone ($159 p<0.354).  Total psychiatric-related healthcare costs were also significantly lower among aripiprazole patients than patients in each of the other treatment groups (p<0.05).  CONCLUSIONS: Among patients with bipolar disorder and newly treated with atypical antipsychotic medications, treatment with aripiprazole was associated with a lower risk of psychiatric hospitalization compared to patients receiving other atypical antipsychotic medications and lower psychiatric-related inpatient costs compared to patients receiving ziprasidone, quetiapine, or olanzapine.  These data are consistent with previous published findings studying the Ingenix integrated claims dataset.  The results of this study suggest that aripiprazole may offer an economic advantage over other atypical antipsychotic medications in this patient population.