Objective To evaluate the prognostic values of CURB-65 score and inflammatory factors in hospitalized patients with community-acquired pneumonia (CAP). Methods A retrospective study was conducted in hospitalized adult CAP patients in West China Hospital between January 1st, and December 31th, 2013. Data of CURB-65 score and serum levels of inflammatory factors (WBC, ESR, PCT, CRP, IL-6 and ALB) on admission and clinical outcomes were collected. The associations between CURB-65 score, inflammatory factors and clinical outcomes were examined. Logistic regression analysis was performed to develop combined models to predict in-hospital death of CAP patients, and ROC analysis was conducted to measure and compare the prognostic values of CURB-65 score, inflammatory factors or combined models. Results A total of 505 hospitalized CAP patients were included. 81 patients died during the hospitalization and the in-hospital mortality rate was 16.0%. Possible risk factors of in-hospital death included old age, male sex, hypertension, cardiovascular or cerebrovascular diseases, multi-lobular pneumonic infiltration, high risk scores, ICU admission, mechanical ventilation and severe pneumonia (all P values<0.05). Logistic regression analysis showed that CURB-65 score, ALB and IL-6 were the independent factors in predicting in-hospital death of CAP patients and the area under curve (AUC) of them while predicting in-hospital death were 0.75 (95%CI 0.69 to 0.81), 0.75 (95%CI 0.69 to 0.81) and 0.75 (95%CI 0.69 to 0.80), respectively. ROC analysis found that ALB and IL-6 could improve the AUC of CURB-65 score significantly while predicting the in-hospital death (P<0.05). When ALB and IL-6 were added to the CURB-65 score simultaneously, the AUC was improved to 0.84 (95%CI 0.80 to 0.87). When IL-6 or ALB was added to the CURB-65 score to form a new scale, the AUC of the new scale was significantly higher than that of the CURB-65 score in predicting in-hospital death (P<0.001). Conclusion The prognostic values of CURB-65 score and inflammatory factors may be not ideal when they are used alone in hospitalized CAP patients. IL-6 and ALB may significantly improve the prognostic value of CURB-65 score in predicting in-hospital death.
The use of clinical predictive modeling to guide clinical decision-making and thus provide accurate diagnosis and treatment services for patients has become a clinical consensus and trend. However, the models available for clinical use are more limited due to unstandardised research methods and poor quality of evidence. This paper introduces the development process of clinical prediction models from six aspects, data collection, model development, performance evaluation, model validation, model presentation and model updating, as well as the clinical prediction model research report statement and risk of bias assessment tools in order to provide methodological references for domestic researchers.