Objective To investigate the correlation of red blood cell distribution width (RDW) to platelet count (PLT) ratio (RPR) with the severity and prognosis of elderly patients with sepsis. Methods A retrospective study was performed on 174 elderly patients with sepsis admitted to the General Hospital of Eastern Theater Command (formerly Nanjing General Hospital of Nanjing Military Command) from January 2019 to June 2020. The general demographic data, results of laboratory and auxiliary examinations, clinical outcomes and other data were collected and analyzed, and the maximum RPR values within 3 d after the diagnosis of sepsis were calculated. SPSS statistics 19.0 was used to perform the statistical analysis. Pearson correlation was employed to analyze the correlation of RPR with acute physiology and chronic health evaluation (APACHE Ⅱ) score, sequential organ failure assessment (SOFA) score, lengths of total hospital stay and ICU stay, and levels of procalcitonin and interleukin-6 (IL-6). According to their clinical outcomes, they were grouped, and non-conditional logistic regression analysis was used to analyze the correlation of RPR with different clinical outcomes of these patients. Receiver operating characteristic (ROC) curve was drawn to evaluate the predictive value of RPR for mortality. Results RPR was positively correlated with APACHE Ⅱ score, SOFA score, lengths of total hospital stay and ICU stay, and levels of procalcitonin and IL-6 (P<0.05). The death group has significantly higher RPR, APACHE Ⅱ score, SOFA score, and incidences of septic shock and severe organ dysfunction than the survival group (P<0.001). After adjustment for age, gender, smoking and drinking status, APACHEⅡ score, and occurrence of septic shock or not, elevated RPR was an independent risk factor for death (OR=3.22,95%CI 1.29-8.01, P=0.012). The area under the ROC curve (AUC) of RPR for predicting death was 0.881 (P<0.001), the best cut-off value was 0.23, sensitivity was 0.905, and specificity was 0.803. The risk of death for those with RPR>0.23 was 9.73 times than those with RPR≤0.23 (OR=9.73,95%CI 1.89-49.99, P=0.006). There was no obvious correlation of RPR with septic shock and severity of organ dysfunction. Conclusion Increased RPR is correlated to the severity and prognosis of elderly patients with sepsis, and is an independent risk factor for predicting death in elderly patients with sepsis.