Objective Fried frailty scale is the very first and most commonly used assessment scale for an operational definition of frailty with its demonstrated success as a predictor of mobility limitations and mortality. However, it is impractical for use in routine clinical practice. We aimed to study whether a simpler modified Fried frailty scale could predict mortality among nursing home residents. Design Retrospective longitudinal follow-up study. Setting Nursing home. Baseline evaluation was performed in 2009. Mortality was assessed after 4 year. Participants Two hundred-twenty-four participants were included. Measurements Residents were assessed for demographic characteristics, falls, dementia, the number of regular medications and chronic diseases, body composition by bioimpedance analysis, basic and instrumental activities of daily living besides frailty status by a modified Fried frailty scale. The residents were assessed for mortality after a median follow-up time of 46 months. The association of frailty with mortality was analyzed by the Kaplan-Meier Log-rank test and multivariate Cox Regression analysis. Results Mortality occurred in 90 (40.2%) of the residents. In multivariate analysis, frailty was an independent predictor of death (Hazzard ratio= 1.4, 95% confidence interval= 1.03-2.6, p=0.03) when adjusted by age, sex, presence of malnutrition, low muscle mass, number of chronic diseases and regular medications. Conclusion Our results suggest that the simpler modified Fried frailty scale can be used as a screening tool for frailty in everyday practice as a tool to identify risky patients for mortality. Future reports studying its role in predicting other adverse outcomes associated with frailty are needed.