Construction and validation of a differentiation scoring system for Spotted fever and Severe fever with thrombocytopenia syndrome

Abstract

Abstract Objective Logistic regression model combined with receiver operating characteristic curve(ROC) was used to construct a clinical differential score model between spotted fever (SF) and severe fever with thrombocytopenia syndrome (SFTS). To provide a reference for clinical preliminary identification. Methods Patients with SF and SFTS from May 2017 to may 2021 in Lu'an secondary hospitals and above were selected, and all patients were confirmed by laboratory. The basic data, epidemic history, clinical data and laboratory data of patients with the two diseases were collected for comparative analysis, logistic regression analysis was conducted to find out the independent influencing factors, and Logistic regression model and scoring system were established. At the same time, the Roc curve was drawn to determine the optimal cut-off value and the area under the ROC curve. Results Multifactorial logistic regression analysis showed that the presence of rash (OR=153.294,,95%CI:7.800-3012.492), elevated C-reactive protein (>10 mg/L) (OR=47.095,95%CI: 3.161-701.735), and normal platelets (>100×109/L ) (OR=37.727, 95% CI: 2.492-571.185) were risk factors for the development of spotted fever, and a score of 1 was assigned to each of these three factors, resulting in a total score of 3 for this scoring system. When the cutoff value was 1, the area under the ROC curve (AUC) of the modeling group was 0.985, the sensitivity was 90.7%, the specificity was 98.8%, the P value in the Hosmer-Lemeshow test was 0.881, and the Kappa value in the consistency test was 0.910. The AUC of validation group was 0.985, the sensitivity was 94.7%, the specificity was 97.1%, the P value of Hosmer-Lemeshow test was 0.977, and the Kappa value of consistency test was 0.918. Conclusion The differential scoring system constructed based on the presence of rash, elevated C-reactive protein, and normal platelets can provide a reference basis for the differentiation of SF from SFTS and has some application value.