Preclinical Identification of Poorly Controlled COPD: Patients with a Single Moderate Exacerbation Matter Too

Abstract

Background and Objectives: Chronic obstructive pulmonary disease (COPD) remains a critical global health challenge, characterized by high morbidity, mortality, and healthcare costs. Current guidelines may overlook patients who present with only one moderate exacerbation or with frequent short-acting beta-agonist (SABA) use. Building on findings from the Seleida study, this research refines the criteria for poor COPD control to include these patients, aiming to improve early identification of high-risk cases in primary care. Methods: A retrospectiveand multicenter study is conducted using data from 110 COPD patients in Spain. Poor control is redefined as having at least one moderate exacerbation or as using three or more SABA inhalers annually. Key predictors, such as SABA/short-acting muscarinic antagonist (SAMA) inhalers and antibiotic prescriptions, are identified using logistic regression and LASSO regularization to enhance predictive accuracy. Results: The model achieves a good predictive performance, with an AUC-ROC of 0.978, sensitivity of 92.86%, and specificity of 87.50%. Key predictors reliably identify high-risk patients, enabling timely interventions. This study demonstrates a statistically significant association between once-daily inhaler therapies and better COPD control compared to multiple daily doses, supported by chi-square analysis (p = 0.008) and binary logistic regression (p = 0.018). Nevertheless, the variable 'daily inhalation frequency' (1 vs. >1 inhalation/day) was excluded from the final model to prevent overfitting. Conclusions: By refining the criteria for COPD control to include patients with at least one moderate exacerbation or frequent SABA use, this model provides a practical tool for early risk stratification in primary care, particularly in resource-limited settings. Early identification of high-risk patients can reduce hospitalizations and healthcare costs, supporting a proactive approach to COPD management. Further validation in larger cohorts is essential to confirm its broader applicability.

Keywords: COPD; SABA use; exacerbations; poor control; predictive model; primary care.

Figure 8

COPD phenotypes according to the number of exacerbations and annual consumption of SABA. COPD, chronic obstructive pulmonary disease; SABA, short-acting β-agonists.

Figure 1

COPD control. COPD, chronic obstructive pulmonary disease.

Figure 2

Contribution of SABA use and exacerbation frequency in the study sample. COPD, chronic obstructive pulmonary disease; SABA, short-acting beta-agonist.

Figure 3

COPD control by treatment regimen. COPD, chronic obstructive pulmonary disease; ICS, inhaled corticosteroids; LABA, long-acting beta-agonists; LAMA, long-acting muscarinic antagonists; SABA, short-acting beta-agonists.

Figure 4

COPD control by daily dose regimen in basal treatment. COPD, chronic obstructive pulmonary disease.

Figure 5

Annual primary care consultations based on the number of COPD exacerbations in the previous year. COPD, chronic obstructive pulmonary disease.

Figure 6

Cross-tabulation of annual prescriptions of antibiotic and oral corticosteroid courses in COPD patients. COPD, chronic obstructive pulmonary disease.

Figure 7

Average annual eosinophil levels in COPD patients by value ranges. COPD, chronic obstructive pulmonary disease.

Figure 9

Violin plot of the key predictive variables in the final COPD control model. (a) Number of SABA and SAMA canisters dispensed in the last year according to the status of disease control; (b) number of antibiotic courses dispensed in patients with good and poor disease control. COPD, chronic obstructive pulmonary disease; SABA, short-acting β-agonists; SAMA, short-acting muscarinic antagonists.