Object of this study was to develop a simple and objective prognostic model for patients with chemotherapy induced febrile neutropenia (FN). A total of 718 patients with FN were prospectively collected from January 2012 to December 2013. We derived our prediction rule using logistic regression with adverse outcome as the primary outcome, and patient demographic and clinical data routinely available at presentation as potential predictor variables. We compared the performance of our model in predicting outcome with that of the Multinational Association for Supportive Care (MASCC) risk score. Risk factors and their points for adverse outcome included in the model were age ≥60 (2), serum procalcitonin ≥0.5 (5), Eastern Cooperative Oncology Group (ECOG) performance score ≥2 (2), Mucositis grade ≥3 (3), initial systolic blood pressure <90 mmHg (3), and initial respiratory rate ≥24/min (3). Total score has its range between 0 and 18 theoretically, however, actual sum of scores ranged from 0 to 15. Our prediction rule stratifies patients with FN into three severity classes, with complication rate of 5.97% in class I (score ≤2), 27.27% in class II (score 3 - 8), and 67.86% in class III (score ≥9). Mortality rate were 0.4% among patients in class I and 17.96% in class III. The area under the curve for our model and the MASCC risk index score were 0.837 (95% CI 0.797 - 0.878) and 0.823 (95% CI 0.795 - 0.87) respectively. We have developed a new prognostic model for FN, through which we can classify patients with FN into 3 classes of increasing adverse outcomes. Further validation of the rule is important before its implementation.