Introduction: Nonischemic dilated cardiomyopathy (NIDCM) is a heterogeneous disease, and patients still have high risk of sudden cardiac death even after receiving treatment. The Selvester QRS score and the ESTIMATED score reported as predictors contain many variables and are complex to calculate. There is a need for a simple predictive score to accurately assess prognosis in clinical practice. Methods: A total of 953 elderly patients (age ≥60 years) diagnosed with NIDCM were enrolled from January 2010 to December 2019. In-hospital and long-term outcomes were studied. Results: Univariate logistic regression analysis showed that the AGEF score was associated with in-hospital mortality (OR: 1.828; 95% CI: 1.559~2.144; p<0.001). Receiver operator characteristic (ROC) curve analysis showed that the AGEF score was excellent at predicting clinical outcomes. The optimal cutoff value of the AGEF score for predicting long-term mortality was 2.50 (AUC=0.743; 95% CI: 0.710~0.776; p<0.001). Kaplan–Meier survival analysis showed that patients with an AGEF score >2.50 had a worse prognosis than those with an AGEF score ≤2.50 (log-rank χ2 103.69, p<0.001). Moreover, multivariate Cox proportional hazard analysis showed that an AGEF score ≤2.50 was associated with a lower risk of long-term mortality in elderly patients with NIDCM (HR: 0.405; 95% CI: 0.310~0.529; p<0.001). Conclusions: The AGEF score could be considered as a simple and useful tool for risk stratification in elderly patients with NIDCM.