The purpose of this prospective study of 2180 consecutive index admissions to an acute geriatric service was to compare in-hospital outcomes of frail older inpatients born in non-English-speaking counties, referred to as culturally and linguistically diverse (CALD) countries in Australia, with those born in English-speaking countries. Multivariate logistic regression was used to model in-hospital mortality and new nursing home placement. Multivariate Cox proportional hazards regression was used to model length of stay. The mean age of all patients was 83 years and 93% were admitted through the emergency department. In multivariate analyses, patients from CALD and non-CALD backgrounds were equally likely to die (CALD odds ratio [OR] 0.69, 95% confidence interval [95% CI] 0.44-1.10) and be newly placed in a nursing home (OR 0.75, 95% CI 0.51-1.12). Patients from CALD backgrounds unable to speak English were more likely to die (11.5% vs. 7.2%, p = 0.02). While patients from CALD backgrounds had significantly shorter lengths of stay in univariate analysis (median 9 days vs. 10 days, p = 0.02), this was not apparent in multivariate analysis (hazard ratio 1.02, 95% CI 0.91-1.14), where the ability to speak English proved to be a strong confounder. While most of the literature shows poorer outcomes of people from minority ethnic groups, our findings indicate that this is not necessarily the case. Developing culturally appropriate services may mitigate some of the adverse outcomes commonly associated with ethnicity. Our findings are particularly relevant to countries populated by multiple ethnic groups.