Predicting the risk of acute respiratory failure among asthma patients—the A2-BEST2 risk score: a retrospective study

• DOI: 10.7717/peerj.16211
• Publication Date: 2023-10-24T08:52:49Z
• Authors: Yanhong Qi; Jing Zhang; Jiaying Lin; Jingwen Yang; Jiangan Guan; Keying Li; Jie Weng; Zhiyi Wang; Chan Chen; Hui Xu
• Keywords: Acute respiratory failure

Objectives Acute respiratory failure (ARF) is a common complication of bronchial asthma (BA). ARF onset increases the risk patient death. This study aims to develop predictive model for in BA patients during hospitalization. Methods was retrospective cohort carried out at two large tertiary hospitals. Three models were developed using three different ways: (1) statistics-driven model, (2) clinical knowledge-driven and (3) decision tree model. The simplest most efficient obtained by comparing their power, stability, practicability. Results included 398 patients, with 298 constituting modeling group 100 validation group. Models A, B, C yielded seven, eleven predictors, respectively. Finally, we chose whose C-statistics Brier scores 0.862 (0.820–0.904) 0.1320, Hosmer-Lemeshow test revealed that this had good calibration. demonstrated satisfactory external internal validation, values 0.890 (0.815–0.965) 0.854 (0.820–0.900), A score incidence created: 2 -BEST Risk Score (A (area pulmonary infection, albumin), BMI, Economic condition, Smoking, T (hormone initiation Time long-term regular medication Treatment)). increased gradually from 1.37% (The ≤ 4) 90.32% ≥ 11.5). Conclusion We constructed seven predictors predict patients. predictor’s simple, practical, supported existing knowledge.