Inhaled Corticosteroid Treatment Regimens and Health Outcomes in a UK COPD Population Study

BACKGROUND: Inhaled corticosteroids (ICS) are a prevailing treatment option for COPD patients but recent guidelines have relegated their use predominantly to patients with frequent exacerbations. Yet large numbers of patients worldwide are currently treated with ICS-containing regimens. We wished to determine in routine clinical practice how common ICS withdrawal is and the differences in health outcomes between patients managed on ICS-containing and non-ICS containing regimens. PATIENTS AND METHODS: COPD patients were identified from the UK primary care electronic healthcare records, between 2014 and 2018. Patients were grouped into three treatment regimens: long-acting beta-agonist (LABA) and inhaled corticosteroids (ICS), LABA and long-acting muscarinic antagonist (LAMA), and triple therapy (LABA, LAMA and ICS). Annual incidence of ICS withdrawal was measured. Multivariable logistic regression was used to identify patient factors associated with withdrawal. Multivariable Poisson regression was used to assess the association of exacerbations and hospitalised pneumonia between the ICS-containing regimens (LABA-ICS and triple therapy) and patients prescribed LABA-LAMA. RESULTS: Of 117,046 patients, around three-quarters were prescribed ICS-containing inhalers but ICS withdrawal occurred annually in only approximately 2–3% of patients. Exacerbations in the past year, but not a past history of pneumonia, were associated with ICS withdrawal. A total of 31,034 patients using three treatment regimens (LABA-ICS, LABA-LAMA or triple therapy) were assessed for their relative risk of exacerbations and pneumonia; the exacerbation risk was slightly lower in LABA-ICS users but the same in triple therapy users, as compared to LABA-LAMA users (LABA-ICS adjusted IRR=0.82 (95% CI 0.73–0.93), triple adjusted IRR=0.99 (95% CI 0.88–1.11)). There was no difference in the pneumonia risk (LABA-ICS adjusted IRR=0.96 (95% CI 0.71–1.31), triple adjusted IRR=1.16 (95% CI 0.87–1.57)). CONCLUSION: Use of ICS-containing inhaled medication is prevalent across the UK while ICS withdrawal from established treatment was relatively uncommon. Exacerbations and pneumonia risk was similar between the ICS-containing and non-ICS containing treatment regimens.