OBJECTIVE: Comparison of frailty instruments in low-middle income countries, where the prevalence of frailty may be higher, is scarce. In addition, less complex diagnostic tools for frailty are important in these settings, especially in acutely ill patients, because of limited time and economic resources. We aimed to compare the performance of 3 frailty instruments for predicting adverse outcomes after 1 year of follow-up in older adults with an acute event or a chronic decompensated disease. DESIGN: Prospective cohort study. SETTING: Geriatric day hospital (GDH) specializing in acute care. PARTICIPANTS: A total of 534 patients (mean age 79.6 ± 8.4 years, 63% female, 64% white) admitted to the GDH. MEASUREMENTS: Frailty was assessed using the Cardiovascular Health Study (CHS) criteria, the Study of Osteoporotic Fracture (SOF) criteria, and the FRAIL (fatigue, resistance, ambulation, illnesses, and loss of weight) questionnaire. Monthly phone contacts were performed over the course of the first year to detect the following outcomes: incident disability, hospitalization, fall, and death. Multivariable Cox proportional hazard regression models were performed to evaluate the association of the outcomes with frailty as defined by the 3 instruments. In addition, we compared the accuracy of these instruments for predicting the outcomes. RESULTS: Prevalence of frailty ranged from 37% (using FRAIL) to 51% (using CHS). After 1 year of follow-up, disability occurred in 33% of the sample, hospitalization in 40%, fall in 44%, and death in 16%. Frailty, as defined by the 3 instruments was associated with all outcomes, whereas prefrailty was associated with disability, using the SOF and FRAIL instruments, and with hospitalization using the CHS and SOF instruments. The accuracy of frailty to predict different outcomes was poor to moderate with area under the curve varying from 0.57 (for fall, with frailty defined by SOF and FRAIL) to 0.69 (for disability, with frailty defined by CHS). CONCLUSIONS: In acutely ill patients from a low-middle income country GDH acute care unit, the CHS, SOF, and FRAIL instruments showed similar performance in predicting adverse outcomes.