Predictors of 30-day pulmonary readmissions in patients with COPD: a case-control study

Abstract

Chronic obstructive pulmonary disease (COPD) exacerbations account for substantial morbidity, mortality, and healthcare expenditure worldwide. Thirty-day readmission for respiratory causes represents a critical quality outcome that reflects the adequacy of acute management and post-discharge care. Data from the Middle East and North Africa (MENA) remain scarce. We conducted a retrospective unmatched case-control study at the American University of Beirut Medical Center (AUBMC), Lebanon (January 2019–August 2025). Cases ( n = 150) were patients readmitted within 30 days of discharge for a respiratory cause following an index COPD exacerbation admission. Controls ( n = 202) were patients not readmitted for a respiratory cause within 30 days. Univariate and multivariable binary logistic regression were used to identify independent predictors. In the primary multivariable model ( n = 135), blood eosinophil count ≥ 300 cells/µL was the strongest independent risk factor for readmission (aOR 38.01, 95% CI 3.51–411.9; p = 0.003), followed by antidepressant use (aOR 4.61, 95% CI 1.28–16.62; p = 0.020) and lower FEV1 (aOR 0.93 per 1%, 95% CI 0.89–0.98; p = 0.005). Inhaler step-up at discharge was the strongest independent protective predictor (aOR 0.06, 95% CI 0.01–0.39; p = 0.003), followed by post-discharge follow-up attendance (aOR 0.16, 95% CI 0.03–0.79; p = 0.025). Key findings were replicated in a sensitivity analysis in the full cohort. Inhaler step-up at discharge and post-discharge follow-up are the strongest modifiable protective factors against 30-day respiratory readmission. Blood eosinophilia identifies a high-risk phenotype specifically prone to recurrent obstructive exacerbations and may represent a target for ICS-guided discharge prescribing. Psychiatric comorbidity is an independent readmission driver not explained by disease severity. A structured discharge bundle of care incorporating these modifiable elements needs prospective validation.