Analysis of lung function differences in patients with varying severities of chronic obstructive pulmonary disease and construction of a risk decision tree prediction model
10.3760/cma.j.cn341190-20241117-01516
- VernacularTitle:不同病情COPD患者肺功能差异分析并构建风险决策树预测模型
- Author:
Qing ZHOU
1
;
Xianrong XU
1
;
Yi WEI
1
Author Information
1. 浙江省立同德医院呼吸与危重症医学科，杭州 310012
- Publication Type:Journal Article
- Keywords:
Pulmonary disease, chronic obstructive;
Respiratory function tests;
Decision trees;
Forecasting;
Risk factors;
Sex factors;
Carbon monoxide
- From:
Chinese Journal of Primary Medicine and Pharmacy
2025;32(9):1281-1286
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To investigate lung function differences in patients with varying severities of chronic obstructive pulmonary disease and construct a risk decision tree prediction model.Methods:A prospective cross-sectional study was conducted among 102 patients with chronic obstructive pulmonary disease (COPD) who received treatment at Tongde Hospital of Zhejiang Province between September 2022 and September 2023. The severity of each patient's condition was evaluated according to the "Guidelines for the Diagnosis and Management of Chronic Obstructive Pulmonary Disease (Revised Version 2021)." All patients were grouped based on the evaluation results. Upon admission, all patients underwent assessments of pulmonary function, ventilation function, blood gas analysis, and laboratory indicators. The baseline data of the patients were statistically analyzed and compared. Risk factors associated with the progression of COPD were analyzed using ordinal regression, and a decision tree model was constructed using the Classification and Regression Trees algorithm. The receiver operating characteristic curve was used to evaluate the predictive value of the decision model for the progression of COPD in patients.Results:The results of the ordinal regression analysis indicated that being female and having a high diffusing capacity of the lungs for carbon monoxide (DLCO) level were protective factors for the development of COPD ( OR < 1, both P < 0.05), while a high concentration of alveolar nitric oxide (CaNO) was a risk factor ( OR > 1, P < 0.05). According to the decision tree, CaNO, DLCO, and female sex were identified as independent risk factors, with CaNO having the most considerable effect. The receiver operating characteristic curve was generated based on the decision tree model. The results showed that the area under the curve of the decision tree model in the moderate group was 0.782, with an optimal cut-off value of 0.314, a specificity of 0.721, a sensitivity of 0.780, and a Youden index of 0.501. The area under the curve of the severe group was 0.845. Conclusions:Sex, DLCO levels, and CaNO levels are factors affecting the development of COPD in patients. The decision tree model based on these factors can effectively predict the risk of COPD progression.
