Objective To develop an objective and precise prognostic model for assessing severity and prognosis in elderly patients with community-acquired pneumonia （CAP） admitted to the emergency department. Methods A retrospective analysis was conducted on elderly patients with CAP admitted to Department of Emergency， Minhang Hospital， Fudan University between Jun 2018 and Dec 2020. With the primary outcome being the 30-day in-hospital mortality rate of elderly CAP patients， four systemic inflammatory response markers， including the neutrophil-to-lymphocyte ratio （NLR）， monocyte-to-lymphocyte ratio （MLR）， platelet-to-lymphocyte ratio （PLR）， and systemic immune-inflammation index （SII） were evaluated using univariate and multivariate Logistic regression analyses. The predictive performance of different scoring systems was compared. Results A total of 421 elderly CAP cases were enrolled. The results of the multivariate Logistic regression analysis demonstrated that NLR was an independent risk factor for elderly inpatients with CAP. We combined NLR with the existing CURB-65 score for joint optimization to construct a scoring system or a clinical prognosis model， by quantifying and assigning optimal cut-off value of 11.4 for NLR， and established the NLR+CURB-65 score. The ROC curve was constructed to compare the areas under the curve of the three different scoring systems （NLR， CURB-65， and NLR+CURB-65）. The area under the curve of the NLR+CURB-65 score was significantly higher than that of the CURB-65 score. Based on the optimal cut-off value of 3 for NLR+CURB-65 score， the patients were stratified into high-risk group （n=188） and low-risk group （n=233）. The K-M survival curve was utilized and indicated that compared with high-risk group， low-risk group had a lower mortality rate and a higher discharge rate. Conclusion For elderly emergency hospitalized patients with CAP， the combination of NLR and CURB-65 score showed high predictive value for assessing disease severity and prognosis.