Development and validation of a predictive model for COPD: a multicenter study.

Area of Science:

• Pulmonary Medicine
• Medical Informatics
• Epidemiology

Background:

• Chronic obstructive pulmonary disease (COPD) is a significant global health concern, ranking as the third leading cause of mortality worldwide.
• COPD presents a substantial public health challenge, particularly in China, necessitating improved diagnostic and management strategies.

Purpose of the Study:

• To develop and validate a predictive model for diagnosing COPD.
• To identify key risk indicators for COPD not previously emphasized in diagnostic criteria.
• To enable risk stratification for patients with COPD.

Main Methods:

• Data from 1,056 inpatients and outpatients (COPD and non-COPD) were collected from January 2018 to December 2022 across three hospitals.
• Logistic regression was employed to construct a predictive model using a training set, with internal and external validation performed on separate datasets.
• The model's performance was assessed using accuracy, sensitivity, specificity, and Area Under the Curve (AUC).

Main Results:

• Six significant risk factors for COPD were identified: age, second-hand smoke exposure, cough, and three distinct patterns of wheezing.
• The developed predictive model demonstrated high performance with 94.1% accuracy, 98.5% sensitivity, and 89.2% specificity in internal validation (AUC=0.976).
• External validation yielded an AUC of 0.691, indicating the model's generalizability.

Conclusions:

• A robust predictive model for COPD diagnosis has been successfully developed.
• The model incorporates key clinical indicators, offering a valuable tool for early detection and risk assessment of COPD.
• The identified risk factors and the predictive equation provide a foundation for targeted interventions and improved patient management strategies.