The potentially inappropriate use of inhaled long-acting beta agonist/corticosteroid (LABA/ICS) combinations in COPD patients for whom this treatment is not recommended has clinical and economic implications.

This retrospective analysis of anonymized electronic medical records in the UK Health Improvement Network (THIN) database was conducted to identify factors associated with step-up from long-acting muscarinic antagonist (LAMA) to LAMA+LABA/ICS therapy. Secondary objectives included time to step-up, Global Initiative for Chronic Obstructive Lung Disease (GOLD) and Medical Research Council (MRC) classification. Data were included from COPD patients between 1 June 2010 and 4 September 2014, aged ≥35 years at first LAMA treatment, with continuous enrolment >360 days before the index event (date of first LAMA prescription) who received LAMA monotherapy only prior to step-up. Time to step-up was analysed using a Cox regression model with time-varying covariates using a stepwise model selection procedure.

Data from 8773 patients (6199 LAMA [136 deaths]; 2438 LAMA+LABA/ICS) were included. Multivariable analysis revealed that exacerbations (composite), elective secondary care contact, markers of COPD proactive planned care, and reactive COPD care within the primary care setting were clinically and statistically significantly associated with step-up. Statistically significant factors negatively associated with step-up were being female and having diabetes (Table). Univariate analysis revealed FEV1, COPD severity and MRC classification to be significant predictors of step-up. These were not included in the multivariable model due to reduced observations, but sensitivity analyses including each in turn confirmed the above predictors. 28% of the cohort received step-up therapy, the majority (23%) within 2 years of LAMA monotherapy initiation. Assessment per GOLD classification suggests that step-up was appropriate in most patients (group A, 18%; B, 21%; C, 26%; D, 35%). Assessment of MRC score (mean, median) in the step-up group (baseline: 2.45, 2.00; follow-up: 2.74, 3.00) suggests that patients who were stepped-up became more symptomatic prior to step-up.

These results show that COPD exacerbations were the most significant predictor of therapy step-up and that patients with initially stable disease are unlikely to require step-up. Therapy step-up appears to be appropriate in the majority of, but not all patients, and may reflect adherence to national guidelines.