Inhaled corticosteroids (ICS) for COPD have been much debated. Our aim was to investigate characteristics of ICS prescribed COPD patients managed only in general practice compared to those also managed in secondary care. Participating general practitioners recruited patients with COPD (ICPC 2nd ed. code R95) currently prescribed ICS (ACT code R03AK and R03BA). Data on demographics, comorbidities, smoking habits, spirometry, dyspnea score and exacerbation history were retrieved from medical records. Logistic regression analysis was applied to detect predictors associated with management in secondary care. 2,279 COPD patients (45% males and mean age 71 years) were recruited in primary care. Compared to patients managed in primary care only (n = 1,179), patients also managed in secondary care (n = 560) were younger (p = 0.013), had lower BMI, more life-time tobacco exposure (p = 0.03), more exacerbations (p < 0.001) and hospitalizations (p < 0.001) and lower FEV₁/FVC-ratio (0.59 versus 0.52, respectively). Compared to patients managed in only primary care, logistic regression analysis revealed that management also in secondary care was associated to MRC-score ≥3 (OR 2.70; 95% CI 1.50–4.86; p = 0.001), FEV₁%pred (OR 0.98; 95% CI 0.95 to 0.99; p = 0.036), and systemic corticosteroids for COPD exacerbation (OR 1.44; 95% CI 1.10–1.89; p = 0.008). In COPD patients prescribed ICS recruited in primary care, patients also managed in secondary care had more respiratory symptoms, lower lung function and exacerbations treated with systemic corticosteroids indicating that the most severe COPD patients, in general, are referred for specialist care.