There are no effective therapies for severe fever with thrombocytopenia syndrome (SFTS), and existing predictors of mortality are still controversial. This retrospective study aimed to identify reliable early-stage indicators for predicting fatal outcomes in 217 patients hospitalized with an SFTS diagnosis between March 2021 and November 2023; 157 of the patients survived, and 60 died. Demographics, clinical characteristics, and laboratory parameters were reassessed in both groups. The mean age of participants was 64.0 (interquartile range: 54.5-71.0) years, and 42.4% (92/217) were males. Based on a multivariate Cox regression analysis, the blood urea nitrogen-to-serum albumin ratio (BAR) (hazard ratio [HR]:4.751; 95% CI: 2.208-10.226; P <0.001), procalcitonin level (HR: 1.946; 95% CI: 1.080-3.507; P = 0.027), and central nervous system symptoms (HR: 3.257; 95% CI, 1.628-6.513; P = 0.001) were independent risk factors for mortality in SFTS patients. According to a receiver operating characteristic curve analysis, a BAR with an area under the curve of 0.913 (95% CI: 0.873-0.953; P <0.001), a sensitivity of 76.7%, and a specificity of 90.4% showed better predictive performance for fatal outcomes than other classical indicators reported. The Kaplan-Meier survival curve confirmed that an increased BAR was linked with an unfavorable prognosis in SFTS patients (P <0.001 by log-rank test). In conclusion, the results indicate that high BAR levels are markedly related to substandard outcomes and are a reliable and readily accessible predictor of fatal outcomes in SFTS patients.