Department of Respiratory and Critical Care Medicine, The First Hospital of China Medical University, Shenyang 110000, People's Republic of China.
Department of Mathematics and Statistics, Xi'an JiaoTong University, Xi'an 710049, People's Republic of China.
Int J Chron Obstruct Pulmon Dis. 2020 Jul 31;15:1849-1861. doi: 10.2147/COPD.S250199. eCollection 2020.
The diagnosis of chronic obstructive pulmonary disease (COPD) is challenging, especially in the primary institution which lacks spirometer. To reduce the rate of COPD missed diagnoses in Northeast China, which has a higher prevalence of COPD, this study aimed to establish efficient primary screening and discriminant models of COPD in this region.
Subjects from Northeast China were enrolled from December 2017 to April 2019 from The First Hospital of China Medical University. Pulmonary function tests and questionnaire were given to all participants. Using illness or no illness as the goal for screening models and disease severity as the goal for discriminant models, multivariate linear regression, logical regression, linear discriminant analysis, K-nearest neighbor, decision tree and support vector machine were constructed through R language and Python software. After comparing effectiveness among them, the most optimal primary screening and discriminant models were established.
Enrolled were 232 COPD patients (124 GOLD I-II and 108 GOLD III-IV) and 218 normal controls. Eight primary screening models were established. The optimal model was Y = -1.2562-0.3891X (education level) + 1.7996X (dyspnea) + 0.5102X (cooking fuel grade) + 1.498X (smoking index) + 0.8077X (family history)-0.5552X (BMI) + 0.538X (cough with sputum) + 2.0328X (wheezing) + 1.3378X (farmers) + 0.8187X (mother's smoking exposure history during pregnancy)-0.389X (kitchen ventilation) + 0.6888X (childhood heating). Six discriminant models were established. The optimal model was decision tree (the optimal variables: dyspnea (x), cooking fuel grade (x), second-hand smoking index (x), BMI (x), cough (x), cough with sputum (x), wheezing (x), farmer (x), kitchen ventilation (x), and childhood heating (x)). The code was established to combine the discriminant model with computer technology.
Many factors were related to COPD in Northeast China. Stepwise logistic regression and decision tree were the optimal screening and discriminant models for COPD in this region.
慢性阻塞性肺疾病（COPD）的诊断具有挑战性，特别是在缺乏肺量计的基层医疗机构。为了降低东北地区 COPD 漏诊率，本研究旨在建立东北地区 COPD 的有效基层筛查和判别模型。
本研究于 2017 年 12 月至 2019 年 4 月在中国医科大学第一附属医院招募了来自中国东北地区的受试者。对所有参与者进行肺功能测试和问卷调查。以患病或不患病作为筛查模型的目标，以疾病严重程度作为判别模型的目标，通过 R 语言和 Python 软件构建多元线性回归、逻辑回归、线性判别分析、K-最近邻、决策树和支持向量机。在比较它们的效果后，建立了最优化的基层筛查和判别模型。
共纳入 232 例 COPD 患者（124 例 GOLD I-II 期和 108 例 GOLD III-IV 期）和 218 例正常对照。建立了 8 种基层筛查模型。最优模型为 Y=-1.2562-0.3891X（教育水平）+1.7996X（呼吸困难）+0.5102X（烹饪燃料等级）+1.498X（吸烟指数）+0.8077X（家族史）-0.5552X（BMI）+0.538X（咳嗽咳痰）+2.0328X（喘息）+1.3378X（农民）+0.8187X（母亲孕期吸烟暴露史）-0.389X（厨房通风）+0.6888X（儿童期采暖）。建立了 6 种判别模型。最优模型为决策树（最优变量：呼吸困难（x）、烹饪燃料等级（x）、二手烟指数（x）、BMI（x）、咳嗽（x）、咳嗽咳痰（x）、喘息（x）、农民（x）、厨房通风（x）和儿童期采暖（x））。建立了结合判别模型和计算机技术的代码。
东北地区 COPD 与多种因素相关。逐步逻辑回归和决策树是该地区 COPD 的最优筛查和判别模型。
