Aims and objectives. The use of maintenance medication in patients with COPD in real life is known to deviate from recommendations in guidelines, which are largely based on randomized controlled trials and selected populations. We used the longitudinal COSYCONET (COPD and Systemic Consequences - Comorbidities Network) cohort to analyze factors linked to the use of COPD drugs under non-interventional circumstances.

Methods. Patients with GOLD grades 0 to 4 participating in visits 1 to 5 were included. Data covered the period from 2010 to 2018. Generalized linear models were used to examine the relation of COPD characteristics to different types of respiratory medication.

Results. A total of 1043 patients were included. The duration of observation was 4.5 years. Use of respiratory medication depended on Global Initiative for Chronic Obstructive Lung Disease (GOLD) grades 0-4 and groups A-D. Long-acting muscarinic antagonist therapy increased over time, and was associated with low CO diffusing capacity, while inhaled corticosteroid (ICS) use decreased. Active smoking was associated with less maintenance therapy in general, and female sex with less ICS use. From the eight items of the COPD Assessment Test (CAT), only hill and stair climbing was consistently linked to treatment.

Conclusions. Using data from a large, close to real-life observational cohort, we identified factors linked to the use of various types of respiratory COPD medication. Overall, use was consistent with GOLD recommendations. Beyond this, we identified other correlates of medication use that may help us to understand and improve therapy decisions in clinical practice.