Evidence from US national health surveys suggests that age-adjusted prevalence of stroke risk factors varies considerably by race/ethnicity. What is not known is whether resource use among hospitalized stroke patients, and the associated cost of inpatient care, differ by race/ethnicity. OBJECTIVE: To assess racial/ethnic differences in hospital length of stay and cost related to inpatient treatment of intracerebral hemorrhage (ICH) in US hospitals. METHODS: The 2002 Health care Cost and Utilization Project database was used to examine short-stay acute-care hospital discharges among adult ICH patients. This dataset includes all discharges from 995 hospitals in 35 states. Patients were identified based on primary ICD-9-CM discharge diagnosis of ICH (431.xx). Patient-level information included demographics (including race/ethnicity: White, Black, Hispanic, Asian), hospital characteristics, comorbidities, primary payer, admission source, discharge destination, DRG, length of stay (LOS), and hospital charges. Costs were estimated by applying hospital-specific cost-to-charge ratios to charges. Adjusted LOS and cost (controlling for age, in-hospital mortality, and other covariates) by race/ethnicity were estimated using multivariate least-squares regression. RESULTS: Black, Hispanic, and Asian patients with a primary diagnosis of ICH at discharge were significantly younger, on average, than White patients (73.5 vs. 61.2, 63.9, and 67.9 years, respectively; p<0.01 for all comparisons). Black and Hispanic patients experienced longer hospital stays (p<0.01) and incurred higher costs (p<0.01), on average, than White patients. Among Black and Hispanic patients, adjusted length of stay (mean costs) per discharge were approximately 2.3 days ($1400) and 1.7 ($3400) higher, respectively, as compared to White patients. Asian patients had longer adjusted stays (2.6 additional days, p<0.01) and higher costs ($830, p<0.31). CONCLUSIONS: There are meaningful differences in length of stay and cost of ICH hospitalizations among patients with different race/ethnicity. Improved acute treatment of stroke in this high-risk population may help to improve overall outcomes in these subgroups.