Few studies have focused on frailty as a predictor of mortality and readmission among inpatients in the acute care setting, especially over long follow-up periods. We conducted this study to determine the impact of the frailty on subsequent mortality and readmission in this setting. This study was a prospective observational study conducted in the acute geriatric wards, with a three-year follow-up duration. We assessed frailty via the 36-item Frailty Index (FI), and a cut-off value of 0.25 was used to identify the presence or absence of frailty. We collected survival and readmission information through telephone interviews at 12, 24, and 36 months. We used the Cox regression model to examine the association between frailty and outcomes interested (death and readmission). The present study included 271 patients (mean age: 81.1 years old; 20.3% females), of whom 21.4% died during the 3-year follow-up period. One hundred and thirty-three patients (49.1%) were identified as being frail. The prevalence of frailty was similar in men and women (46.8% vs.58.2%, P = 0.130). Compared with non-frail patients, death and hospital readmission rates of frail patients were increased. Frailty was an independent predictor of 3-year death (adjusted hazard ratio (HR): 2.09; 95% confidence interval (CI): 1.20 to 3.63) and readmission (adjusted HR: 1.40; 95% CI: 1.04 to 1.88) after adjusting for several potential confounders. Frailty is prevalent among older inpatients and is a valuable predictor of 3-year mortality and hospital readmission in an acute care setting.