Development and Validation of a Method to Estimate COPD Severity in Multiple Datasets: A Retrospective Study.

Area of Science:

• Pulmonary Medicine
• Health Informatics
• Epidemiology

Background:

• Chronic obstructive pulmonary disease (COPD) severity impacts patient outcomes, including symptoms, hospitalizations, and mortality.
• Electronic medical record heterogeneity challenges accurate COPD severity assessment across different regions.
• The Global Initiative for Chronic Obstructive Lung Disease (GOLD) 2011 classification provides a framework for categorizing COPD severity.

Purpose of the Study:

• To develop and validate a method for approximating COPD severity using the GOLD 2011 classification.
• To address challenges in measuring COPD severity due to incomplete electronic medical record data.
• To enable more consistent COPD severity assessment across diverse geographical settings.

Main Methods:

• Retrospective analysis of patient data from multiple international databases (CPRD, IMRD, Disease Analyzer, Longitudinal Patient Data).
• Development of the method using patients with complete data for GOLD 2011 group calculation in the CPRD.
• Validation using ordinal logistic models at COPD diagnosis and index (first triple therapy) in the full study population.

Main Results:

• The developed method correctly classified 74.4% and 75.9% of patients into severe and non-severe COPD categories at diagnosis and index, respectively.
• Key covariates included age, gender, time to triple therapy, healthcare resource use, comorbidities, and prescriptions.
• The study included 4,579 patients at diagnosis and 12,539 patients at index.

Conclusions:

• A validated method was developed to approximate COPD severity based on GOLD 2011 categories.
• This approach can be applied to patients lacking all parameters required for direct GOLD classification.
• The method facilitates more standardized COPD severity assessment in real-world data.